Release 0.3.11 (#835)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

.eslintrc.cjs

@@ -0,0 +1,76 @@
+module.exports = {
+  root: true,
+  extends: [
+    "turbo",
+    "prettier",
+    "plugin:@typescript-eslint/recommended-type-checked-only",
+  ],
+  parserOptions: {
+    project: true,
+    __tsconfigRootDir: __dirname,
+  },
+  settings: {
+    react: {
+      version: "999.999.999",
+    },
+  },
+  rules: {
+    "max-params": ["error", 4],
+    "prefer-const": "error",
+    "@typescript-eslint/no-floating-promises": [
+      "error",
+      {
+        ignoreIIFE: true,
+      },
+    ],
+    "@typescript-eslint/await-thenable": "off",
+    "@typescript-eslint/ban-ts-comment": "off",
+    "@typescript-eslint/ban-types": "off",
+    "no-array-constructor": "off",
+    "@typescript-eslint/no-array-constructor": "off",
+    "@typescript-eslint/no-base-to-string": "off",
+    "@typescript-eslint/no-duplicate-enum-values": "off",
+    "@typescript-eslint/no-duplicate-type-constituents": "off",
+    "@typescript-eslint/no-explicit-any": "off",
+    "@typescript-eslint/no-extra-non-null-assertion": "off",
+    "@typescript-eslint/no-for-in-array": "off",
+    "no-implied-eval": "off",
+    "@typescript-eslint/no-implied-eval": "off",
+    "no-loss-of-precision": "off",
+    "@typescript-eslint/no-loss-of-precision": "off",
+    "@typescript-eslint/no-misused-new": "off",
+    "@typescript-eslint/no-misused-promises": "off",
+    "@typescript-eslint/no-namespace": "off",
+    "@typescript-eslint/no-non-null-asserted-optional-chain": "off",
+    "@typescript-eslint/no-redundant-type-constituents": "off",
+    "@typescript-eslint/no-this-alias": "off",
+    "@typescript-eslint/no-unnecessary-type-assertion": "off",
+    "@typescript-eslint/no-unnecessary-type-constraint": "off",
+    "@typescript-eslint/no-unsafe-argument": "off",
+    "@typescript-eslint/no-unsafe-assignment": "off",
+    "@typescript-eslint/no-unsafe-call": "off",
+    "@typescript-eslint/no-unsafe-declaration-merging": "off",
+    "@typescript-eslint/no-unsafe-enum-comparison": "off",
+    "@typescript-eslint/no-unsafe-member-access": "off",
+    "@typescript-eslint/no-unsafe-return": "off",
+    "no-unused-vars": "off",
+    "@typescript-eslint/no-unused-vars": "off",
+    "@typescript-eslint/no-var-requires": "off",
+    "@typescript-eslint/prefer-as-const": "off",
+    "require-await": "off",
+    "@typescript-eslint/require-await": "off",
+    "@typescript-eslint/restrict-plus-operands": "off",
+    "@typescript-eslint/restrict-template-expressions": "off",
+    "@typescript-eslint/triple-slash-reference": "off",
+    "@typescript-eslint/unbound-method": "off",
+  },
+  overrides: [
+    {
+      files: ["examples/**/*.ts"],
+      rules: {
+        "turbo/no-undeclared-env-vars": "off",
+      },
+    },
+  ],
+  ignorePatterns: ["dist/", "lib/", "deps/"],
+};

.eslintrc.js

@@ -1,23 +0,0 @@
-module.exports = {
-  root: true,
-  // This tells ESLint to load the config from the package `eslint-config-custom`
-  extends: ["custom"],
-  settings: {
-    next: {
-      rootDir: ["apps/*/"],
-    },
-  },
-  rules: {
-    "max-params": ["error", 4],
-    "prefer-const": "error",
-  },
-  overrides: [
-    {
-      files: ["examples/**/*.ts"],
-      rules: {
-        "turbo/no-undeclared-env-vars": "off",
-      },
-    },
-  ],
-  ignorePatterns: ["dist/", "lib/"],
-};

.github/workflows/lint_on_push_or_pull.yml

@@ -13,9 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
-        with:
-          version: latest
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

.github/workflows/publish.yml

@@ -14,7 +14,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

.github/workflows/publish_github.yml

@@ -0,0 +1,37 @@
+name: Publish to GitHub Releases
+
+on:
+  push:
+    tags:
+      - "llamaindex@*"
+
+jobs:
+  build-and-publish:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Repo
+        uses: actions/checkout@v4
+
+      - uses: pnpm/action-setup@v3
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
+
+      - name: Install dependencies
+        run: pnpm install
+
+      - name: Build tarball
+        run: |
+          pnpm pack
+        working-directory: packages/core
+
+      - name: Create release
+        uses: ncipollo/release-action@v1
+        with:
+          artifacts: "packages/core/llamaindex-*.tgz"
+          name: Release ${{ github.ref }}
+          bodyFile: "packages/core/CHANGELOG.md"
+          token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/release.yml

@@ -0,0 +1,57 @@
+name: Release
+
+on:
+  push:
+    branches:
+      - main
+
+concurrency: ${{ github.workflow }}-${{ github.ref }}
+
+jobs:
+  release:
+    name: Release
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Repo
+        uses: actions/checkout@v4
+
+      - uses: pnpm/action-setup@v3
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
+
+      - name: Install dependencies
+        run: pnpm install
+
+      - name: Add auth token to .npmrc file
+        run: |
+          cat << EOF >> ".npmrc"
+            //registry.npmjs.org/:_authToken=$NPM_TOKEN
+          EOF
+        env:
+          NPM_TOKEN: ${{ secrets.NPM_TOKEN }}
+
+      - name: Get changeset status
+        id: get-changeset-status
+        run: |
+          pnpm changeset status --output .changeset/status.json
+          new_version=$(jq -r '.releases[] | select(.name == "llamaindex") | .newVersion' < .changeset/status.json)
+          rm -v .changeset/status.json
+          echo "new-version=${new_version}" >> "$GITHUB_OUTPUT"
+
+      - name: Create Release Pull Request or Publish to npm
+        id: changesets
+        uses: changesets/action@v1
+        with:
+          commit: Release ${{ steps.get-changeset-status.outputs.new-version }}
+          title: Release ${{ steps.get-changeset-status.outputs.new-version }}
+          # update version PR with the latest changesets
+          version: pnpm new-version
+          # build package and call changeset publish
+          publish: pnpm release
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          NPM_TOKEN: ${{ secrets.NPM_TOKEN }}

.github/workflows/test.yml

@@ -1,18 +1,55 @@
 name: Run Tests
 
-on: [push, pull_request]
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
 
 jobs:
+  e2e:
+    strategy:
+      fail-fast: false
+      matrix:
+        node-version: [18.x, 20.x, 22.x]
+    name: E2E on Node.js ${{ matrix.node-version }}
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: pnpm/action-setup@v3
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: ${{ matrix.node-version }}
+          cache: "pnpm"
+      - name: Install dependencies
+        run: pnpm install
+      - name: Run E2E Tests
+        run: pnpm run e2e
+
   test:
+    strategy:
+      fail-fast: false
+      matrix:
+        node-version: [18.x, 20.x, 22.x]
+    name: Test on Node.js ${{ matrix.node-version }}
     runs-on: ubuntu-latest
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
-          node-version-file: ".nvmrc"
+          node-version: ${{ matrix.node-version }}
           cache: "pnpm"
       - name: Install dependencies
         run: pnpm install
@@ -23,7 +60,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -32,24 +69,33 @@ jobs:
       - name: Install dependencies
         run: pnpm install
       - name: Build
-        run: pnpm run build --filter llamaindex
+        run: pnpm run build
+      - name: Use Build For Examples
+        run: pnpm link ../packages/core/
+        working-directory: ./examples
       - name: Run Type Check
         run: pnpm run type-check
       - name: Run Circular Dependency Check
-        run: pnpm run circular-check
-        working-directory: ./packages/core
+        run: pnpm dlx turbo run circular-check
       - uses: actions/upload-artifact@v3
         if: failure()
         with:
           name: typecheck-build-dist
           path: ./packages/core/dist
           if-no-files-found: error
-  core-edge-runtime:
+  e2e-core-examples:
+    strategy:
+      matrix:
+        packages:
+          - cloudflare-worker-agent
+          - nextjs-agent
+          - nextjs-edge-runtime
+          - waku-query-engine
     runs-on: ubuntu-latest
-
+    name: Build Core Example (${{ matrix.packages }})
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -57,17 +103,18 @@ jobs:
           cache: "pnpm"
       - name: Install dependencies
         run: pnpm install
-      - name: Build
-        run: pnpm run build --filter @llamaindex/edge
-      - name: Build Edge Runtime
+      - name: Build llamaindex
         run: pnpm run build
-        working-directory: ./packages/edge/e2e/test-edge-runtime
+      - name: Build ${{ matrix.packages }}
+        run: pnpm run build
+        working-directory: packages/core/e2e/examples/${{ matrix.packages }}
+
   typecheck-examples:
     runs-on: ubuntu-latest
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v2
+      - uses: pnpm/action-setup@v3
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -76,7 +123,7 @@ jobs:
       - name: Install dependencies
         run: pnpm install
       - name: Build
-        run: pnpm run build --filter llamaindex
+        run: pnpm run build
       - name: Copy examples
         run: rsync -rv --exclude=node_modules ./examples ${{ runner.temp }}
       - name: Pack @llamaindex/env

.husky/pre-push

@@ -1 +0,0 @@
-pnpm test

.npmrc

@@ -1,2 +1,5 @@
 auto-install-peers = true
 enable-pre-post-scripts = true
+prefer-workspace-packages = true
+save-workspace-protocol = true
+link-workspace-packages = true

.nvmrc

@@ -1 +1 @@
-18
+20

.vscode/launch.json

@@ -10,8 +10,9 @@
       "name": "Debug Example",
       "skipFiles": ["<node_internals>/**"],
       "runtimeExecutable": "pnpm",
+      "console": "integratedTerminal",
       "cwd": "${workspaceFolder}/examples",
-      "runtimeArgs": ["ts-node", "${fileBasename}"]
+      "runtimeArgs": ["npx", "tsx", "${file}"]
     }
   ]
 }

CONTRIBUTING.md

@@ -91,16 +91,10 @@ Please send a descriptive changeset for each PR.
 
 ## Publishing (maintainers only)
 
-To publish a new version of the library, first create a new version:
+The [Release Github Action](.github/workflows/release.yml) is automatically generating and updating a
+PR called "Release {version}".
 
-```shell
-pnpm new-version
-```
+This PR will update the `package.json` and `CHANGELOG.md` files of each package according to
+the current changesets in the [.changeset](.changeset/) folder.
 
-If everything looks good, commit the generated files and release the new version:
-
-```shell
-pnpm release
-git push # push to the main branch
-git push --tags
-```
+If this PR is merged it will automatically add version tags to the repository and publish the updated packages to NPM.

README-turborepo.md

@@ -1,81 +0,0 @@
-# Turborepo starter
-
-This is an official starter Turborepo.
-
-## Using this example
-
-Run the following command:
-
-```sh
-npx create-turbo@latest
-```
-
-## What's inside?
-
-This Turborepo includes the following packages/apps:
-
-### Apps and Packages
-
-- `docs`: a [Next.js](https://nextjs.org/) app
-- `web`: another [Next.js](https://nextjs.org/) app
-- `ui`: a stub React component library shared by both `web` and `docs` applications
-- `eslint-config-custom`: `eslint` configurations (includes `eslint-config-next` and `eslint-config-prettier`)
-- `tsconfig`: `tsconfig.json`s used throughout the monorepo
-
-Each package/app is 100% [TypeScript](https://www.typescriptlang.org/).
-
-### Utilities
-
-This Turborepo has some additional tools already setup for you:
-
-- [TypeScript](https://www.typescriptlang.org/) for static type checking
-- [ESLint](https://eslint.org/) for code linting
-- [Prettier](https://prettier.io) for code formatting
-
-### Build
-
-To build all apps and packages, run the following command:
-
-```
-cd my-turborepo
-pnpm build
-```
-
-### Develop
-
-To develop all apps and packages, run the following command:
-
-```
-cd my-turborepo
-pnpm dev
-```
-
-### Remote Caching
-
-Turborepo can use a technique known as [Remote Caching](https://turbo.build/repo/docs/core-concepts/remote-caching) to share cache artifacts across machines, enabling you to share build caches with your team and CI/CD pipelines.
-
-By default, Turborepo will cache locally. To enable Remote Caching you will need an account with Vercel. If you don't have an account you can [create one](https://vercel.com/signup), then enter the following commands:
-
-```
-cd my-turborepo
-npx turbo login
-```
-
-This will authenticate the Turborepo CLI with your [Vercel account](https://vercel.com/docs/concepts/personal-accounts/overview).
-
-Next, you can link your Turborepo to your Remote Cache by running the following command from the root of your Turborepo:
-
-```
-npx turbo link
-```
-
-## Useful Links
-
-Learn more about the power of Turborepo:
-
-- [Tasks](https://turbo.build/repo/docs/core-concepts/monorepos/running-tasks)
-- [Caching](https://turbo.build/repo/docs/core-concepts/caching)
-- [Remote Caching](https://turbo.build/repo/docs/core-concepts/remote-caching)
-- [Filtering](https://turbo.build/repo/docs/core-concepts/monorepos/filtering)
-- [Configuration Options](https://turbo.build/repo/docs/reference/configuration)
-- [CLI Usage](https://turbo.build/repo/docs/reference/command-line-reference)

README.md

@@ -19,25 +19,29 @@ Try examples online:
 
 LlamaIndex.TS aims to be a lightweight, easy to use set of libraries to help you integrate large language models into your applications with your own data.
 
-## Getting started with an example:
+## Multiple JS Environment Support
 
-LlamaIndex.TS requires Node v18 or higher. You can download it from https://nodejs.org or use https://nvm.sh (our preferred option).
+LlamaIndex.TS supports multiple JS environments, including:
 
-In a new folder:
+- Node.js (18, 20, 22) ✅
+- Deno ✅
+- Bun ✅
+- React Server Components (Next.js) ✅
 
-```bash
-export OPENAI_API_KEY="sk-......" # Replace with your key from https://platform.openai.com/account/api-keys
-pnpm init
-pnpm install typescript
-pnpm exec tsc --init # if needed
+For now, browser support is limited due to the lack of support for [AsyncLocalStorage-like APIs](https://github.com/tc39/proposal-async-context)
+
+## Getting started
+
+```shell
+npm install llamaindex
 pnpm install llamaindex
-pnpm install @types/node
+yarn add llamaindex
+jsr install @llamaindex/core
 ```
 
-Create the file example.ts
+### Node.js
 
 ```ts
-// example.ts
 import fs from "fs/promises";
 import { Document, VectorStoreIndex } from "llamaindex";
 
@@ -67,10 +71,121 @@ async function main() {
 main();
 ```
 
-Then you can run it using
-
 ```bash
-pnpm dlx ts-node example.ts
+# `pnpm install tsx` before running the script
+node --import tsx ./main.ts
+```
+
+### Next.js
+
+First, you will need to add a llamaindex plugin to your Next.js project.
+
+```js
+// next.config.js
+const withLlamaIndex = require("llamaindex/next");
+
+module.exports = withLlamaIndex({
+  // your next.js config
+});
+```
+
+You can combine `ai` with `llamaindex` in Next.js with RSC (React Server Components).
+
+```tsx
+// src/apps/page.tsx
+"use client";
+import { chatWithAgent } from "@/actions";
+import type { JSX } from "react";
+import { useFormState } from "react-dom";
+
+// You can use the Edge runtime in Next.js by adding this line:
+// export const runtime = "edge";
+
+export default function Home() {
+  const [ui, action] = useFormState<JSX.Element | null>(async () => {
+    return chatWithAgent("hello!", []);
+  }, null);
+  return (
+    <main>
+      {ui}
+      <form action={action}>
+        <button>Chat</button>
+      </form>
+    </main>
+  );
+}
+```
+
+```tsx
+// src/actions/index.ts
+"use server";
+import { createStreamableUI } from "ai/rsc";
+import { OpenAIAgent } from "llamaindex";
+import type { ChatMessage } from "llamaindex/llm/types";
+
+export async function chatWithAgent(
+  question: string,
+  prevMessages: ChatMessage[] = [],
+) {
+  const agent = new OpenAIAgent({
+    tools: [
+      // ... adding your tools here
+    ],
+  });
+  const responseStream = await agent.chat({
+    stream: true,
+    message: question,
+    chatHistory: prevMessages,
+  });
+  const uiStream = createStreamableUI(<div>loading...</div>);
+  responseStream
+    .pipeTo(
+      new WritableStream({
+        start: () => {
+          uiStream.update("response:");
+        },
+        write: async (message) => {
+          uiStream.append(message.response.delta);
+        },
+      }),
+    )
+    .catch(console.error);
+  return uiStream.value;
+}
+```
+
+### Cloudflare Workers
+
+```ts
+// src/index.ts
+export default {
+  async fetch(
+    request: Request,
+    env: Env,
+    ctx: ExecutionContext,
+  ): Promise<Response> {
+    const { setEnvs } = await import("@llamaindex/env");
+    // set environment variables so that the OpenAIAgent can use them
+    setEnvs(env);
+    const { OpenAIAgent } = await import("llamaindex");
+    const agent = new OpenAIAgent({
+      tools: [],
+    });
+    const responseStream = await agent.chat({
+      stream: true,
+      message: "Hello? What is the weather today?",
+    });
+    const textEncoder = new TextEncoder();
+    const response = responseStream.pipeThrough(
+      new TransformStream({
+        transform: (chunk, controller) => {
+          controller.enqueue(textEncoder.encode(chunk.response.delta));
+        },
+      }),
+    );
+    return new Response(response);
+  },
+};
 ```
 
 ## Playground
@@ -93,72 +208,25 @@ Check out our NextJS playground at https://llama-playground.vercel.app/. The sou
 
 - [SimplePrompt](/packages/core/src/Prompt.ts): A simple standardized function call definition that takes in inputs and formats them in a template literal. SimplePrompts can be specialized using currying and combined using other SimplePrompt functions.
 
-## Using NextJS
+## Tips when using in non-Node.js environments
 
-If you're using the NextJS App Router, you can choose between the Node.js and the [Edge runtime](https://nextjs.org/docs/app/building-your-application/rendering/edge-and-nodejs-runtimes#edge-runtime).
+When you are importing `llamaindex` in a non-Node.js environment(such as React Server Components, Cloudflare Workers, etc.)
+Some classes are not exported from top-level entry file.
 
-With NextJS 13 and 14, using the Node.js runtime is the default. You can explicitly set the Edge runtime in your [router handler](https://nextjs.org/docs/app/building-your-application/routing/route-handlers) by adding this line:
+The reason is that some classes are only compatible with Node.js runtime,(e.g. `PDFReader`) which uses Node.js specific APIs(like `fs`, `child_process`, `crypto`).
+
+If you need any of those classes, you have to import them instead directly though their file path in the package.
+Here's an example for importing the `PineconeVectorStore` class:
 
 ```typescript
-export const runtime = "edge";
+import { PineconeVectorStore } from "llamaindex/storage/vectorStore/PineconeVectorStore";
 ```
 
-The following sections explain further differences in using the Node.js or Edge runtime.
-
-### Using the Node.js runtime
-
-Add the following config to your `next.config.js` to ignore specific packages in the server-side bundling:
-
-```js
-// next.config.js
-/** @type {import('next').NextConfig} */
-const nextConfig = {
-  experimental: {
-    serverComponentsExternalPackages: ["pdf2json", "@zilliz/milvus2-sdk-node"],
-  },
-  webpack: (config) => {
-    config.resolve.alias = {
-      ...config.resolve.alias,
-      sharp$: false,
-      "onnxruntime-node$": false,
-    };
-    return config;
-  },
-};
-
-module.exports = nextConfig;
-```
-
-### Using the Edge runtime
-
-We publish a dedicated package (`@llamaindex/edge` instead of `llamaindex`) for using the Edge runtime. To use it, first install the package:
-
-```shell
-pnpm install @llamaindex/edge
-```
-
-> _Note_: Ensure that your `package.json` doesn't include the `llamaindex` package if you're using `@llamaindex/edge`.
-
-Then make sure to use the correct import statement in your code:
+As the `PDFReader` is not working with the Edge runtime, here's how to use the `SimpleDirectoryReader` with the `LlamaParseReader` to load PDFs:
 
 ```typescript
-// replace 'llamaindex' with '@llamaindex/edge'
-import {} from "@llamaindex/edge";
-```
-
-A further difference is that the `@llamaindex/edge` package doesn't export classes from the `readers` or `storage` folders. The reason is that most of these classes are not compatible with the Edge runtime.
-
-If you need any of those classes, you have to import them instead directly. Here's an example for importing the `PineconeVectorStore` class:
-
-```typescript
-import { PineconeVectorStore } from "@llamaindex/edge/storage/vectorStore/PineconeVectorStore";
-```
-
-As the `PDFReader` is not with the Edge runtime, here's how to use the `SimpleDirectoryReader` with the `LlamaParseReader` to load PDFs:
-
-```typescript
-import { SimpleDirectoryReader } from "@llamaindex/edge/readers/SimpleDirectoryReader";
-import { LlamaParseReader } from "@llamaindex/edge/readers/LlamaParseReader";
+import { SimpleDirectoryReader } from "llamaindex/readers/SimpleDirectoryReader";
+import { LlamaParseReader } from "llamaindex/readers/LlamaParseReader";
 
 export const DATA_DIR = "./data";
 
@@ -176,14 +244,14 @@ export async function getDocuments() {
 
 > _Note_: Reader classes have to be added explictly to the `fileExtToReader` map in the Edge version of the `SimpleDirectoryReader`.
 
-You'll find a complete example of using the Edge runtime with LlamaIndexTS here: https://github.com/run-llama/create_llama_projects/tree/main/nextjs-edge-llamaparse
+You'll find a complete example with LlamaIndexTS here: https://github.com/run-llama/create_llama_projects/tree/main/nextjs-edge-llamaparse
 
 ## Supported LLMs:
 
 - OpenAI GPT-3.5-turbo and GPT-4
 - Anthropic Claude 3 (Opus, Sonnet, and Haiku) and the legacy models (Claude 2 and Instant)
 - Groq LLMs
-- Llama2 Chat LLMs (70B, 13B, and 7B parameters)
+- Llama2/3 Chat LLMs (70B, 13B, and 7B parameters)
 - MistralAI Chat LLMs
 - Fireworks Chat LLMs
 

apps/docs/CHANGELOG.md

@@ -1,5 +1,125 @@
 # docs
 
+## 0.0.19
+
+### Patch Changes
+
+- Updated dependencies [e072c45]
+- Updated dependencies [9e133ac]
+- Updated dependencies [447105a]
+- Updated dependencies [320be3f]
+  - llamaindex@0.3.11
+
+## 0.0.18
+
+### Patch Changes
+
+- Updated dependencies [4aba02e]
+  - llamaindex@0.3.10
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [c3747d0]
+  - llamaindex@0.3.9
+
+## 0.0.16
+
+### Patch Changes
+
+- Updated dependencies [ce94780]
+  - llamaindex@0.3.8
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [b6a6606]
+- Updated dependencies [b6a6606]
+  - llamaindex@0.3.7
+
+## 0.0.14
+
+### Patch Changes
+
+- Updated dependencies [efa326a]
+  - llamaindex@0.3.6
+
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [bc7a11c]
+- Updated dependencies [2fe2b81]
+- Updated dependencies [5596e31]
+- Updated dependencies [e74fe88]
+- Updated dependencies [be5df5b]
+  - llamaindex@0.3.5
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [1dce275]
+- Updated dependencies [d10533e]
+- Updated dependencies [2008efe]
+- Updated dependencies [5e61934]
+- Updated dependencies [9e74a43]
+- Updated dependencies [ee719a1]
+  - llamaindex@0.3.4
+
+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [e8c41c5]
+  - llamaindex@0.3.3
+
+## 0.0.10
+
+### Patch Changes
+
+- Updated dependencies [61103b6]
+  - llamaindex@0.3.2
+
+## 0.0.9
+
+### Patch Changes
+
+- Updated dependencies [46227f2]
+  - llamaindex@0.3.1
+
+## 0.0.8
+
+### Patch Changes
+
+- Updated dependencies [5016f21]
+  - llamaindex@0.3.0
+
+## 0.0.7
+
+### Patch Changes
+
+- Updated dependencies [6277105]
+  - llamaindex@0.2.13
+
+## 0.0.6
+
+### Patch Changes
+
+- Updated dependencies [d8d952d]
+  - llamaindex@0.2.12
+
+## 0.0.5
+
+### Patch Changes
+
+- Updated dependencies [87142b2]
+- Updated dependencies [5a6cc0e]
+- Updated dependencies [87142b2]
+  - llamaindex@0.2.11
+
 ## 0.0.4
 
 ### Patch Changes

apps/docs/blog/2024-04-26-v0.3.0.md

@@ -0,0 +1,493 @@
+---
+title: LlamaIndexTS v0.3.0
+description: This is my first post on Docusaurus.
+slug: welcome-llamaindexts-v0.3
+authors:
+  - name: Alex Yang
+    title: LlamaIndexTS maintainer, Node.js Member
+    url: https://github.com/himself65
+    image_url: https://github.com/himself65.png
+tags: [llamaindex, agent]
+hide_table_of_contents: false
+---
+
+- [What's new in LlamaIndexTS v0.3.0](#whats-new-in-llamaindexts-v030)
+- [Improvement in LlamaIndexTS v0.3.0](#improvement-in-llamaindexts-v030)
+- [What's the next?](#whats-the-next)
+
+## What's new in LlamaIndexTS v0.3.0
+
+## Agents
+
+In this release, we've not only ported the Agent module from the LlamaIndex Python version but have significantly
+enhanced it to be more powerful and user-friendly for JavaScript/TypeScript applications.
+
+Starting from v0.3.0, we are introducing multiple agents specifically designed for RAG applications, including:
+
+- `OpenAIAgent`
+- `AnthropicAgent`
+- `ReActAgent`:
+
+```ts
+import { OpenAIAgent } from "llamaindex";
+import { tools } from "./tools";
+
+const agent = new OpenAIAgent({
+  tools: [...tools],
+});
+const { response } = await agent.chat({
+  message: "What is weather today?",
+  stream: false,
+});
+
+console.log(response.message.content);
+```
+
+We are also introducing the abstract AgentRunner class, which allows you to create your own agent by simply implementing
+the task handler.
+
+```ts
+import { AgentRunner, OpenAI } from "llamaindex";
+
+class MyLLM extends OpenAI {}
+
+export class MyAgentWorker extends AgentWorker<MyLLM> {
+  taskHandler = MyAgent.taskHandler;
+}
+
+export class MyAgent extends AgentRunner<MyLLM> {
+  constructor(params: Params) {
+    super({
+      llm: params.llm,
+      chatHistory: params.chatHistory ?? [],
+      systemPrompt: params.systemPrompt ?? null,
+      runner: new MyAgentWorker(),
+      tools:
+        "tools" in params
+          ? params.tools
+          : params.toolRetriever.retrieve.bind(params.toolRetriever),
+    });
+  }
+
+  // create store is a function to create a store for each task, by default it only includes `messages` and `toolOutputs`
+  createStore = AgentRunner.defaultCreateStore;
+
+  static taskHandler: TaskHandler<Anthropic> = async (step, enqueueOutput) => {
+    const { llm, stream } = step.context;
+    // initialize the input
+    const response = await llm.chat({
+      stream,
+      messages: step.context.store.messages,
+    });
+    // store the response for next task step
+    step.context.store.messages = [
+      ...step.context.store.messages,
+      response.message,
+    ];
+    // your logic here to decide whether to continue the task
+    const shouldContinue = Math.random(); /* <-- replace with your logic here */
+    enqueueOutput({
+      taskStep: step,
+      output: response,
+      isLast: !shouldContinue,
+    });
+    if (shouldContinue) {
+      const content = await someHeavyFunctionCall();
+      // if you want to continue the task, you can insert your new context for the next task step
+      step.context.store.messages = [
+        ...step.context.store.messages,
+        {
+          content,
+          role: "user",
+        },
+      ];
+    }
+  };
+}
+```
+
+### Web Stream API for Streaming response
+
+Web Stream is a web standard utilized in many modern web frameworks and libraries (like React 19, Deno, Node 22). We
+have migrated streaming responses to Web Stream to ensure broader compatibility.
+
+For instance, you can use the streaming response in a simple HTTP Server:
+
+```ts
+import { createServer } from "http";
+import { OpenAIAgent } from "llamaindex";
+import { OpenAIStream, streamToResponse } from "ai";
+import { tools } from "./tools";
+
+const agent = new OpenAIAgent({
+  tools: [...tools],
+});
+
+const server = createServer(async (req, res) => {
+  const response = await agent.chat({
+    message: "What is weather today?",
+    stream: true,
+  });
+
+  // Transform the response into a string readable stream
+  const stream: ReadableStream<string> = response.pipeThrough(
+    new TransformStream({
+      transform: (chunk, controller) => {
+        controller.enqueue(chunk.response.delta);
+      },
+    }),
+  );
+
+  // Pipe the stream to the response
+  streamToResponse(stream, res);
+});
+
+server.listen(3000);
+```
+
+Or it can be integrated into React Server Components (RSC) in Next.js:
+
+```tsx
+// app/actions/index.tsx
+"use server";
+import { createStreamableUI } from "ai/rsc";
+import { OpenAIAgent } from "llamaindex";
+import type { ChatMessage } from "llamaindex/llm/types";
+
+export async function chatWithAgent(
+  question: string,
+  prevMessages: ChatMessage[] = [],
+) {
+  const agent = new OpenAIAgent({
+    tools: [],
+  });
+  const responseStream = await agent.chat({
+    stream: true,
+    message: question,
+    chatHistory: prevMessages,
+  });
+  const uiStream = createStreamableUI(<div>loading...</div>);
+  responseStream
+    .pipeTo(
+      new WritableStream({
+        start: () => {
+          uiStream.update("response:");
+        },
+        write: async (message) => {
+          uiStream.append(message.response.delta);
+        },
+      }),
+    )
+    .catch(uiStream.error);
+  return uiStream.value;
+}
+```
+
+```tsx
+// app/src/page.tsx
+"use client";
+import { chatWithAgent } from "@/actions";
+import type { JSX } from "react";
+import { useFormState } from "react-dom";
+
+export const runtime = "edge";
+
+export default function Home() {
+  const [state, action] = useFormState<JSX.Element | null>(async () => {
+    return chatWithAgent("hello!", []);
+  }, null);
+  return (
+    <main>
+      {state}
+      <form action={action}>
+        <button>Chat</button>
+      </form>
+    </main>
+  );
+}
+```
+
+## Improvement in LlamaIndexTS v0.3.0
+
+### Better TypeScript support
+
+We have made significant improvements to the type system to ensure that all code is thoroughly checked before it is
+published. This ongoing enhancement has already resulted in better module reliability and developer experience.
+
+For example, we have improved `FunctionTool` type with generic support:
+
+```ts
+type Input = {
+  a: number;
+  b: number;
+};
+
+const sumNumbers = FunctionTool.from<Input>(
+  ({ a, b }) => `${a + b}`, // a and b will be checked as number
+  // JSON schema will be an error if you type wrong.
+  {
+    name: "sumNumbers",
+    description: "Use this function to sum two numbers",
+    parameters: {
+      type: "object",
+      properties: {
+        a: {
+          type: "number",
+          description: "The first number",
+        },
+        b: {
+          type: "number",
+          description: "The second number",
+        },
+      },
+      required: ["a", "b"],
+    },
+  },
+);
+```
+
+![type checking](./img/function_tool_example.png)
+
+### Better Next.js, Deno, Cloudflare Worker, and Waku(Vite) support
+
+In addition to Node.js, LlamaIndexTS now offers enhanced support for Next.js, Deno, and Cloudflare Workers, making it
+more versatile across different platforms.
+
+For now, you can install llamaindex and directly import it into your existing Next.js, Deno or Cloudflare Worker project
+**without any extra configuration**.
+
+#### [Deno](https://deno.com/)
+
+You can use LlamaIndexTS in Deno by installation through JSR:
+
+```sh
+jsr add @llamaindex/core
+```
+
+#### [Cloudflare Worker](https://developers.cloudflare.com/workers/)
+
+For Cloudflare Workers, here is a starter template:
+
+```typescript
+export default {
+  async fetch(
+    request: Request,
+    env: Env,
+    ctx: ExecutionContext,
+  ): Promise<Response> {
+    const { setEnvs } = await import("@llamaindex/env");
+    setEnvs(env);
+    const { OpenAIAgent } = await import("llamaindex");
+    const agent = new OpenAIAgent({
+      tools: [],
+    });
+    const responseStream = await agent.chat({
+      stream: true,
+      message: "Hello? What is the weather today?",
+    });
+    const textEncoder = new TextEncoder();
+    const response = responseStream.pipeThrough(
+      new TransformStream({
+        transform: (chunk, controller) => {
+          controller.enqueue(textEncoder.encode(chunk.response.delta));
+        },
+      }),
+    );
+    return new Response(response);
+  },
+};
+```
+
+### [Waku (Vite)](https://waku.gg/)
+
+Waku powered by Vite is a minimal React framework that supports multiple JS environments, including Deno, Cloudflare, and
+Node.js.
+
+You can use LlamaIndexTS with Node.js output to enable full Node.js support with React.
+
+```sh
+npm install llamaindex
+```
+
+```ts
+// file: src/actions.ts
+"use server";
+import { Document, VectorStoreIndex } from "llamaindex";
+import { readFile } from "node:fs/promises";
+
+const path = "node_modules/llamaindex/examples/abramov.txt";
+
+const essay = await readFile(path, "utf-8");
+
+// Create Document object with essay
+const document = new Document({ text: essay, id_: path });
+
+// Split text and create embeddings. Store them in a VectorStoreIndex
+const index = await VectorStoreIndex.fromDocuments([document]);
+
+const queryEngine = index.asQueryEngine();
+
+export async function chatWithAI(question: string): Promise<string> {
+  const { response } = await queryEngine.query({ query: question });
+  return response;
+}
+```
+
+```tsx
+// file: src/pages/index.tsx
+import { chatWithAI } from "./actions";
+
+export default async function HomePage() {
+  return (
+    <div>
+      <Chat askQuestion={chatWithAI} />
+    </div>
+  );
+}
+```
+
+```tsx
+// file: src/components/Chat.tsx
+"use client";
+
+export type ChatProps = {
+  askQuestion: (question: string) => Promise<string>;
+};
+
+export const Chat = (props: ChatProps) => {
+  const [response, setResponse] = useState<string | null>(null);
+
+  return (
+    <section className="border-blue-400 -mx-4 mt-4 rounded border border-dashed p-4">
+      <h2 className="text-lg font-bold">Chat with AI</h2>
+      {response ? (
+        <p className="text-sm text-gray-600 max-w-sm">{response}</p>
+      ) : null}
+      <form
+        action={async (formData) => {
+          const question = formData.get("question") as string | null;
+          if (question) {
+            setResponse(await props.askQuestion(question));
+          }
+        }}
+      >
+        <input
+          type="text"
+          name="question"
+          className="border border-gray-400 rounded-sm px-2 py-0.5 text-sm"
+        />
+        <button className="rounded-sm bg-black px-2 py-0.5 text-sm text-white">
+          Ask
+        </button>
+      </form>
+    </section>
+  );
+};
+```
+
+```shell
+waku dev # development mode
+waku build # build for production
+waku start # start the production server
+```
+
+Note that not all the modules are supported in all JS environments because of
+lack of the file system, network API,
+and incompatibility with the Node.js API by upstream dependencies.
+
+But we are trying to make it more compatible with all the environments.
+
+## What's the next?
+
+As we continue to develop LlamaIndexTS, our focus remains on providing more comprehensive and powerful tools for
+creating custom agents.
+
+### Align with the Python `llama-index`
+
+We aim to align LlamaIndexTS with the Python version to ensure API consistency and ease of use for developers familiar
+with the Python ecosystem.
+
+### Align with the Web Standard and JS development
+
+Not all python APIs are compatible and easy to use in JavaScript/TypeScript.
+We are trying to make the API more compatible with the Web Standard and JavaScript modern development.
+
+### More Agents
+
+Future releases will introduce more agents from the Python Llama-Index and explore APIs tailored to real-world use
+cases.
+
+### 🧪 `@llamaindex/tool`
+
+We are exploring innovative ways to create tools for agents. The `@llamaindex/tool` library allows you to transform any
+function into a tool for an agent, simplifying the development process and reducing runtime costs.
+
+```ts
+export function getWeather(city: string) {
+  return `The weather in ${city} is sunny.`;
+}
+
+// you don't need to worry about the shcema with different llm tools
+export function getTemperature(city: string) {
+  return `The temperature in ${city} is 25°C.`;
+}
+
+export function getCurrentCity() {
+  return "New York";
+}
+```
+
+These functions can be easily integrated into your applications, such as Next.js:
+
+```ts
+"use server";
+import { OpenAI } from "openai";
+import { getTools } from "@llamaindex/tool";
+
+export async function chat(message: string) {
+  const openai = new OpenAI();
+  openai.chat.completions.create({
+    messages: [
+      {
+        role: "user",
+        content: "What is the weather in the current city?",
+      },
+    ],
+    tools: getTools("openai"),
+  });
+}
+```
+
+```ts
+// next.config.js
+const withTool = require("@llamaindex/tool/next");
+
+const config = {
+  // Your original Next.js config
+};
+module.exports = withTool(config);
+```
+
+The functions are automatically transformed into tools for the agent at compile time, which eliminates any extra runtime
+costs. This feature is particularly beneficial when you need to debug or deploy your assistant.
+
+For deploying your local functions into OpenAI, you can use a simple command:
+
+```sh
+npm install -g @llamaindex/tool
+mkai --tools ./src/index.llama.ts
+# Successfully created assistant: asst_XXX
+# chat with your assistant by `chatai --assistant asst_XXX`
+chatai --assistant asst_XXX
+# Open your browser and chat with your assistant
+# Running at http://localhost:3000
+```
+
+This deployment process simplifies the testing and implementation of your custom tools in a live environment.
+
+As this project is still in its early stages, we continue to explore the best ways to create and integrate tools for
+agents. For more information and updates, visit the @llamaindex/tool repository.
+
+This release of LlamaIndexTS v0.3.0 marks a significant step forward in our journey to provide developers with robust,
+flexible tools for building advanced agents. We are excited to see how our community utilizes these new capabilities to
+create innovative solutions and look forward to continuing to support and enhance LlamaIndexTS in future updates.

apps/docs/docs/examples/agent.mdx

@@ -4,82 +4,7 @@ A built-in agent that can take decisions and reasoning based on the tools provid
 
 ## OpenAI Agent
 
-```ts
-import { FunctionTool, OpenAIAgent } from "llamaindex";
+import CodeBlock from "@theme/CodeBlock";
+import CodeSource from "!raw-loader!../../../../examples/agent/openai";
 
-// Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-
-// Define the parameters of the sum function as a JSON schema
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-// Define the parameters of the divide function as a JSON schema
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The dividend to divide",
-    },
-    b: {
-      type: "number",
-      description: "The divisor to divide by",
-    },
-  },
-  required: ["a", "b"],
-};
-
-async function main() {
-  // Create a function tool from the sum function
-  const sumFunctionTool = new FunctionTool(sumNumbers, {
-    name: "sumNumbers",
-    description: "Use this function to sum two numbers",
-    parameters: sumJSON,
-  });
-
-  // Create a function tool from the divide function
-  const divideFunctionTool = new FunctionTool(divideNumbers, {
-    name: "divideNumbers",
-    description: "Use this function to divide two numbers"
-    parameters: divideJSON,
-  });
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [sumFunctionTool, divideFunctionTool],
-    verbose: true,
-  });
-
-  // Chat with the agent
-  const response = await agent.chat({
-    message: "How much is 5 + 5? then divide by 2",
-  });
-
-  // Print the response
-  console.log(String(response));
-}
-
-main().then(() => {
-  console.log("Done");
-});
-```
+<CodeBlock language="ts">{CodeSource}</CodeBlock>

apps/docs/docs/modules/agent/index.md

@@ -11,4 +11,10 @@ An “agent” is an automated reasoning and decision engine. It takes in a user
 
 LlamaIndex.TS comes with a few built-in agents, but you can also create your own. The built-in agents include:
 
-- [OpenAI Agent](./openai.mdx)
+- OpenAI Agent
+- Anthropic Agent
+- ReACT Agent
+
+## Examples
+
+- [OpenAI Agent](../../examples/agent.mdx)

apps/docs/docs/modules/agent/multi_document_agent.mdx

@@ -1,309 +0,0 @@
-# Multi-Document Agent
-
-In this guide, you learn towards setting up an agent that can effectively answer different types of questions over a larger set of documents.
-
-These questions include the following
-
-- QA over a specific doc
-- QA comparing different docs
-- Summaries over a specific doc
-- Comparing summaries between different docs
-
-We do this with the following architecture:
-
-- setup a “document agent” over each Document: each doc agent can do QA/summarization within its doc
-- setup a top-level agent over this set of document agents. Do tool retrieval and then do CoT over the set of tools to answer a question.
-
-## Setup and Download Data
-
-We first start by installing the necessary libraries and downloading the data.
-
-```bash
-pnpm i llamaindex
-```
-
-```ts
-import {
-  Document,
-  ObjectIndex,
-  OpenAI,
-  OpenAIAgent,
-  QueryEngineTool,
-  SimpleNodeParser,
-  SimpleToolNodeMapping,
-  SummaryIndex,
-  VectorStoreIndex,
-  Settings,
-  storageContextFromDefaults,
-} from "llamaindex";
-```
-
-And then for the data we will run through a list of countries and download the wikipedia page for each country.
-
-```ts
-import fs from "fs";
-import path from "path";
-
-const dataPath = path.join(__dirname, "tmp_data");
-
-const extractWikipediaTitle = async (title: string) => {
-  const fileExists = fs.existsSync(path.join(dataPath, `${title}.txt`));
-
-  if (fileExists) {
-    console.log(`File already exists for the title: ${title}`);
-    return;
-  }
-
-  const queryParams = new URLSearchParams({
-    action: "query",
-    format: "json",
-    titles: title,
-    prop: "extracts",
-    explaintext: "true",
-  });
-
-  const url = `https://en.wikipedia.org/w/api.php?${queryParams}`;
-
-  const response = await fetch(url);
-  const data: any = await response.json();
-
-  const pages = data.query.pages;
-  const page = pages[Object.keys(pages)[0]];
-  const wikiText = page.extract;
-
-  await new Promise((resolve) => {
-    fs.writeFile(path.join(dataPath, `${title}.txt`), wikiText, (err: any) => {
-      if (err) {
-        console.error(err);
-        resolve(title);
-        return;
-      }
-      console.log(`${title} stored in file!`);
-
-      resolve(title);
-    });
-  });
-};
-```
-
-```ts
-export const extractWikipedia = async (titles: string[]) => {
-  if (!fs.existsSync(dataPath)) {
-    fs.mkdirSync(dataPath);
-  }
-
-  for await (const title of titles) {
-    await extractWikipediaTitle(title);
-  }
-
-  console.log("Extration finished!");
-```
-
-These files will be saved in the `tmp_data` folder.
-
-Now we can call the function to download the data for each country.
-
-```ts
-await extractWikipedia([
-  "Brazil",
-  "United States",
-  "Canada",
-  "Mexico",
-  "Argentina",
-  "Chile",
-  "Colombia",
-  "Peru",
-  "Venezuela",
-  "Ecuador",
-  "Bolivia",
-  "Paraguay",
-  "Uruguay",
-  "Guyana",
-  "Suriname",
-  "French Guiana",
-  "Falkland Islands",
-]);
-```
-
-## Load the data
-
-Now that we have the data, we can load it into the LlamaIndex and store as a document.
-
-```ts
-import { Document } from "llamaindex";
-
-const countryDocs: Record<string, Document> = {};
-
-for (const title of wikiTitles) {
-  const path = `./agent/helpers/tmp_data/${title}.txt`;
-  const text = await fs.readFile(path, "utf-8");
-  const document = new Document({ text: text, id_: path });
-  countryDocs[title] = document;
-}
-```
-
-## Setup LLM and StorageContext
-
-We will be using gpt-4 for this example and we will use the `StorageContext` to store the documents in-memory.
-
-```ts
-Settings.llm = new OpenAI({
-  model: "gpt-4",
-});
-
-const storageContext = await storageContextFromDefaults({
-  persistDir: "./storage",
-});
-```
-
-## Building Multi-Document Agents
-
-In this section we show you how to construct the multi-document agent. We first build a document agent for each document, and then define the top-level parent agent with an object index.
-
-```ts
-const documentAgents: Record<string, any> = {};
-const queryEngines: Record<string, any> = {};
-```
-
-Now we iterate over each country and create a document agent for each one.
-
-### Build Agent for each Document
-
-In this section we define “document agents” for each document.
-
-We define both a vector index (for semantic search) and summary index (for summarization) for each document. The two query engines are then converted into tools that are passed to an OpenAI function calling agent.
-
-This document agent can dynamically choose to perform semantic search or summarization within a given document.
-
-We create a separate document agent for each coutnry.
-
-```ts
-for (const title of wikiTitles) {
-  // parse the document into nodes
-  const nodes = new SimpleNodeParser({
-    chunkSize: 200,
-    chunkOverlap: 20,
-  }).getNodesFromDocuments([countryDocs[title]]);
-
-  // create the vector index for specific search
-  const vectorIndex = await VectorStoreIndex.init({
-    storageContext: storageContext,
-    nodes,
-  });
-
-  // create the summary index for broader search
-  const summaryIndex = await SummaryIndex.init({
-    nodes,
-  });
-
-  const vectorQueryEngine = summaryIndex.asQueryEngine();
-  const summaryQueryEngine = summaryIndex.asQueryEngine();
-
-  // create the query engines for each task
-  const queryEngineTools = [
-    new QueryEngineTool({
-      queryEngine: vectorQueryEngine,
-      metadata: {
-        name: "vector_tool",
-        description: `Useful for questions related to specific aspects of ${title} (e.g. the history, arts and culture, sports, demographics, or more).`,
-      },
-    }),
-    new QueryEngineTool({
-      queryEngine: summaryQueryEngine,
-      metadata: {
-        name: "summary_tool",
-        description: `Useful for any requests that require a holistic summary of EVERYTHING about ${title}. For questions about more specific sections, please use the vector_tool.`,
-      },
-    }),
-  ];
-
-  // create the document agent
-  const agent = new OpenAIAgent({
-    tools: queryEngineTools,
-    llm,
-    verbose: true,
-  });
-
-  documentAgents[title] = agent;
-  queryEngines[title] = vectorIndex.asQueryEngine();
-}
-```
-
-## Build Top-Level Agent
-
-Now we define the top-level agent that can answer questions over the set of document agents.
-
-This agent takes in all document agents as tools. This specific agent RetrieverOpenAIAgent performs tool retrieval before tool use (unlike a default agent that tries to put all tools in the prompt).
-
-Here we use a top-k retriever, but we encourage you to customize the tool retriever method!
-
-Firstly, we create a tool for each document agent
-
-```ts
-const allTools: QueryEngineTool[] = [];
-```
-
-```ts
-for (const title of wikiTitles) {
-  const wikiSummary = `
-    This content contains Wikipedia articles about ${title}.
-    Use this tool if you want to answer any questions about ${title}
-  `;
-
-  const docTool = new QueryEngineTool({
-    queryEngine: documentAgents[title],
-    metadata: {
-      name: `tool_${title}`,
-      description: wikiSummary,
-    },
-  });
-
-  allTools.push(docTool);
-}
-```
-
-Our top level agent will use this document agents as tools and use toolRetriever to retrieve the best tool to answer a question.
-
-```ts
-// map the tools to nodes
-const toolMapping = SimpleToolNodeMapping.fromObjects(allTools);
-
-// create the object index
-const objectIndex = await ObjectIndex.fromObjects(
-  allTools,
-  toolMapping,
-  VectorStoreIndex,
-  {
-    storageContext,
-  },
-);
-
-// create the top agent
-const topAgent = new OpenAIAgent({
-  toolRetriever: await objectIndex.asRetriever({}),
-  llm,
-  verbose: true,
-  prefixMessages: [
-    {
-      content:
-        "You are an agent designed to answer queries about a set of given countries. Please always use the tools provided to answer a question. Do not rely on prior knowledge.",
-      role: "system",
-    },
-  ],
-});
-```
-
-## Use the Agent
-
-Now we can use the agent to answer questions.
-
-```ts
-const response = await topAgent.chat({
-  message: "Tell me the differences between Brazil and Canada economics?",
-});
-
-// print output
-console.log(response);
-```
-
-You can find the full code for this example [here](https://github.com/run-llama/LlamaIndexTS/tree/main/examples/agent/multi-document-agent.ts)

apps/docs/docs/modules/agent/openai.mdx

@@ -1,187 +0,0 @@
----
-sidebar_position: 0
----
-
-# OpenAI Agent
-
-OpenAI API that supports function calling, it’s never been easier to build your own agent!
-
-In this notebook tutorial, we showcase how to write your own OpenAI agent
-
-## Setup
-
-First, you need to install the `llamaindex` package. You can do this by running the following command in your terminal:
-
-```bash
-pnpm i llamaindex
-```
-
-Then we can define a function to sum two numbers and another function to divide two numbers.
-
-```ts
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-```
-
-## Create a function tool
-
-Now we can create a function tool from the sum function and another function tool from the divide function.
-
-For the parameters of the sum function, we can define a JSON schema.
-
-### JSON Schema
-
-```ts
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The dividend a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The divisor b to divide by",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const sumFunctionTool = new FunctionTool(sumNumbers, {
-  name: "sumNumbers",
-  description: "Use this function to sum two numbers",
-  parameters: sumJSON,
-});
-
-const divideFunctionTool = new FunctionTool(divideNumbers, {
-  name: "divideNumbers",
-  description: "Use this function to divide two numbers",
-  parameters: divideJSON,
-});
-```
-
-## Create an OpenAIAgent
-
-Now we can create an OpenAIAgent with the function tools.
-
-```ts
-const agent = new OpenAIAgent({
-  tools: [sumFunctionTool, divideFunctionTool],
-  verbose: true,
-});
-```
-
-## Chat with the agent
-
-Now we can chat with the agent.
-
-```ts
-const response = await agent.chat({
-  message: "How much is 5 + 5? then divide by 2",
-});
-
-console.log(String(response));
-```
-
-## Full code
-
-```ts
-import { FunctionTool, OpenAIAgent } from "llamaindex";
-
-// Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-
-// Define the parameters of the sum function as a JSON schema
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-// Define the parameters of the divide function as a JSON schema
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The argument a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The argument b to divide",
-    },
-  },
-  required: ["a", "b"],
-};
-
-async function main() {
-  // Create a function tool from the sum function
-  const sumFunctionTool = new FunctionTool(sumNumbers, {
-    name: "sumNumbers",
-    description: "Use this function to sum two numbers",
-    parameters: sumJSON,
-  });
-
-  // Create a function tool from the divide function
-  const divideFunctionTool = new FunctionTool(divideNumbers, {
-    name: "divideNumbers",
-    description: "Use this function to divide two numbers",
-    parameters: divideJSON,
-  });
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [sumFunctionTool, divideFunctionTool],
-    verbose: true,
-  });
-
-  // Chat with the agent
-  const response = await agent.chat({
-    message: "How much is 5 + 5? then divide by 2",
-  });
-
-  // Print the response
-  console.log(String(response));
-}
-
-main().then(() => {
-  console.log("Done");
-});
-```

apps/docs/docs/modules/agent/query_engine_tool.mdx

@@ -1,132 +0,0 @@
----
-sidebar_position: 1
----
-
-# OpenAI Agent + QueryEngineTool
-
-QueryEngineTool is a tool that allows you to query a vector index. In this example, we will create a vector index from a set of documents and then create a QueryEngineTool from the vector index. We will then create an OpenAIAgent with the QueryEngineTool and chat with the agent.
-
-## Setup
-
-First, you need to install the `llamaindex` package. You can do this by running the following command in your terminal:
-
-```bash
-pnpm i llamaindex
-```
-
-Then you can import the necessary classes and functions.
-
-```ts
-import {
-  OpenAIAgent,
-  SimpleDirectoryReader,
-  VectorStoreIndex,
-  QueryEngineTool,
-} from "llamaindex";
-```
-
-## Create a vector index
-
-Now we can create a vector index from a set of documents.
-
-```ts
-// Load the documents
-const documents = await new SimpleDirectoryReader().loadData({
-  directoryPath: "node_modules/llamaindex/examples/",
-});
-
-// Create a vector index from the documents
-const vectorIndex = await VectorStoreIndex.fromDocuments(documents);
-```
-
-## Create a QueryEngineTool
-
-Now we can create a QueryEngineTool from the vector index.
-
-```ts
-// Create a query engine from the vector index
-const abramovQueryEngine = vectorIndex.asQueryEngine();
-
-// Create a QueryEngineTool with the query engine
-const queryEngineTool = new QueryEngineTool({
-  queryEngine: abramovQueryEngine,
-  metadata: {
-    name: "abramov_query_engine",
-    description: "A query engine for the Abramov documents",
-  },
-});
-```
-
-## Create an OpenAIAgent
-
-```ts
-// Create an OpenAIAgent with the query engine tool tools
-
-const agent = new OpenAIAgent({
-  tools: [queryEngineTool],
-  verbose: true,
-});
-```
-
-## Chat with the agent
-
-Now we can chat with the agent.
-
-```ts
-const response = await agent.chat({
-  message: "What was his salary?",
-});
-
-console.log(String(response));
-```
-
-## Full code
-
-```ts
-import {
-  OpenAIAgent,
-  SimpleDirectoryReader,
-  VectorStoreIndex,
-  QueryEngineTool,
-} from "llamaindex";
-
-async function main() {
-  // Load the documents
-  const documents = await new SimpleDirectoryReader().loadData({
-    directoryPath: "node_modules/llamaindex/examples/",
-  });
-
-  // Create a vector index from the documents
-  const vectorIndex = await VectorStoreIndex.fromDocuments(documents);
-
-  // Create a query engine from the vector index
-  const abramovQueryEngine = vectorIndex.asQueryEngine();
-
-  // Create a QueryEngineTool with the query engine
-  const queryEngineTool = new QueryEngineTool({
-    queryEngine: abramovQueryEngine,
-    metadata: {
-      name: "abramov_query_engine",
-      description: "A query engine for the Abramov documents",
-    },
-  });
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [queryEngineTool],
-    verbose: true,
-  });
-
-  // Chat with the agent
-  const response = await agent.chat({
-    message: "What was his salary?",
-  });
-
-  // Print the response
-  console.log(String(response));
-}
-
-main().then(() => {
-  console.log("Done");
-});
-```

apps/docs/docs/modules/agent/react_agent.mdx

@@ -1,203 +0,0 @@
-# ReAct Agent
-
-The ReAct agent is an AI agent that can reason over the next action, construct an action command, execute the action, and repeat these steps in an iterative loop until the task is complete.
-
-In this notebook tutorial, we showcase how to write your ReAct agent using the `llamaindex` package.
-
-## Setup
-
-First, you need to install the `llamaindex` package. You can do this by running the following command in your terminal:
-
-```bash
-pnpm i llamaindex
-```
-
-And then you can import the `OpenAIAgent` and `FunctionTool` from the `llamaindex` package.
-
-```ts
-import { FunctionTool, OpenAIAgent } from "llamaindex";
-```
-
-Then we can define a function to sum two numbers and another function to divide two numbers.
-
-```ts
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-```
-
-## Create a function tool
-
-Now we can create a function tool from the sum function and another function tool from the divide function.
-
-For the parameters of the sum function, we can define a JSON schema.
-
-### JSON Schema
-
-```ts
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The dividend a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The divisor b to divide by",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const sumFunctionTool = new FunctionTool(sumNumbers, {
-  name: "sumNumbers",
-  description: "Use this function to sum two numbers",
-  parameters: sumJSON,
-});
-
-const divideFunctionTool = new FunctionTool(divideNumbers, {
-  name: "divideNumbers",
-  description: "Use this function to divide two numbers",
-  parameters: divideJSON,
-});
-```
-
-## Create an ReAct
-
-Now we can create an OpenAIAgent with the function tools.
-
-```ts
-const agent = new ReActAgent({
-  tools: [sumFunctionTool, divideFunctionTool],
-  verbose: true,
-});
-```
-
-## Chat with the agent
-
-Now we can chat with the agent.
-
-```ts
-const response = await agent.chat({
-  message: "How much is 5 + 5? then divide by 2",
-});
-
-console.log(String(response));
-```
-
-The output will be:
-
-```bash
-Thought: I need to use a tool to help me answer the question.
-Action: sumNumbers
-Action Input: {"a":5,"b":5}
-
-Observation: 10
-Thought: I can answer without using any more tools.
-Answer: The sum of 5 and 5 is 10, and when divided by 2, the result is 5.
-
-The sum of 5 and 5 is 10, and when divided by 2, the result is 5.
-```
-
-## Full code
-
-```ts
-import { FunctionTool, ReActAgent } from "llamaindex";
-
-// Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-
-// Define the parameters of the sum function as a JSON schema
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-// Define the parameters of the divide function as a JSON schema
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The argument a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The argument b to divide",
-    },
-  },
-  required: ["a", "b"],
-};
-
-async function main() {
-  // Create a function tool from the sum function
-  const sumFunctionTool = new FunctionTool(sumNumbers, {
-    name: "sumNumbers",
-    description: "Use this function to sum two numbers",
-    parameters: sumJSON,
-  });
-
-  // Create a function tool from the divide function
-  const divideFunctionTool = new FunctionTool(divideNumbers, {
-    name: "divideNumbers",
-    description: "Use this function to divide two numbers",
-    parameters: divideJSON,
-  });
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [sumFunctionTool, divideFunctionTool],
-    verbose: true,
-  });
-
-  // Chat with the agent
-  const response = await agent.chat({
-    message: "I want to sum 5 and 5 and then divide by 2",
-  });
-
-  // Print the response
-  console.log(String(response));
-}
-
-main().then(() => {
-  console.log("Done");
-});
-```

apps/docs/docs/modules/embeddings/available_embeddings/gemini.md

@@ -0,0 +1,33 @@
+# Gemini
+
+To use Gemini embeddings, you need to import `GeminiEmbedding` from `llamaindex`.
+
+```ts
+import { GeminiEmbedding, Settings } from "llamaindex";
+
+// Update Embed Model
+Settings.embedModel = new GeminiEmbedding();
+
+const document = new Document({ text: essay, id_: "essay" });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+
+const queryEngine = index.asQueryEngine();
+
+const query = "What is the meaning of life?";
+
+const results = await queryEngine.query({
+  query,
+});
+```
+
+Per default, `GeminiEmbedding` is using the `gemini-pro` model. You can change the model by passing the `model` parameter to the constructor.
+For example:
+
+```ts
+import { GEMINI_MODEL, GeminiEmbedding } from "llamaindex";
+
+Settings.embedModel = new GeminiEmbedding({
+  model: GEMINI_MODEL.GEMINI_PRO_LATEST,
+});
+```

apps/docs/docs/modules/embeddings/available_embeddings/jinaai.md

@@ -0,0 +1,21 @@
+# Jina AI
+
+To use Jina AI embeddings, you need to import `JinaAIEmbedding` from `llamaindex`.
+
+```ts
+import { JinaAIEmbedding, Settings } from "llamaindex";
+
+Settings.embedModel = new JinaAIEmbedding();
+
+const document = new Document({ text: essay, id_: "essay" });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+
+const queryEngine = index.asQueryEngine();
+
+const query = "What is the meaning of life?";
+
+const results = await queryEngine.query({
+  query,
+});
+```

apps/docs/docs/modules/embeddings/available_embeddings/ollama.md

@@ -1,11 +1,19 @@
 # Ollama
 
-To use Ollama embeddings, you need to import `Ollama` from `llamaindex`.
+To use Ollama embeddings, you need to import `OllamaEmbedding` from `llamaindex`.
+
+Note that you need to pull the embedding model first before using it.
+
+In the example below, we're using the [`nomic-embed-text`](https://ollama.com/library/nomic-embed-text) model, so you have to call:
+
+```shell
+ollama pull nomic-embed-text
+```
 
 ```ts
-import { Ollama, Settings } from "llamaindex";
+import { OllamaEmbedding, Settings } from "llamaindex";
 
-Settings.embedModel = new Ollama();
+Settings.embedModel = new OllamaEmbedding({ model: "nomic-embed-text" });
 
 const document = new Document({ text: essay, id_: "essay" });
 

apps/docs/docs/modules/llms/available_llms/gemini.md

@@ -0,0 +1,71 @@
+# Gemini
+
+## Usage
+
+```ts
+import { Gemini, Settings, GEMINI_MODEL } from "llamaindex";
+
+Settings.llm = new Gemini({
+  model: GEMINI_MODEL.GEMINI_PRO,
+});
+```
+
+## Load and index documents
+
+For this example, we will use a single document. In a real-world scenario, you would have multiple documents to index.
+
+```ts
+const document = new Document({ text: essay, id_: "essay" });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+```
+
+## Query
+
+```ts
+const queryEngine = index.asQueryEngine();
+
+const query = "What is the meaning of life?";
+
+const results = await queryEngine.query({
+  query,
+});
+```
+
+## Full Example
+
+```ts
+import {
+  Gemini,
+  Document,
+  VectorStoreIndex,
+  Settings,
+  GEMINI_MODEL,
+} from "llamaindex";
+
+Settings.llm = new Gemini({
+  model: GEMINI_MODEL.GEMINI_PRO,
+});
+
+async function main() {
+  const document = new Document({ text: essay, id_: "essay" });
+
+  // Load and index documents
+  const index = await VectorStoreIndex.fromDocuments([document]);
+
+  // Create a query engine
+  const queryEngine = index.asQueryEngine({
+    retriever,
+  });
+
+  const query = "What is the meaning of life?";
+
+  // Query
+  const response = await queryEngine.query({
+    query,
+  });
+
+  // Log the response
+  console.log(response.response);
+}
+```

apps/docs/docs/modules/llms/available_llms/llama2.md

@@ -3,7 +3,7 @@
 ## Usage
 
 ```ts
-import { Ollama, Settings } from "llamaindex";
+import { Ollama, Settings, DeuceChatStrategy } from "llamaindex";
 
 Settings.llm = new LlamaDeuce({ chatStrategy: DeuceChatStrategy.META });
 ```
@@ -11,7 +11,12 @@ Settings.llm = new LlamaDeuce({ chatStrategy: DeuceChatStrategy.META });
 ## Usage with Replication
 
 ```ts
-import { Ollama, ReplicateSession, Settings } from "llamaindex";
+import {
+  Ollama,
+  ReplicateSession,
+  Settings,
+  DeuceChatStrategy,
+} from "llamaindex";
 
 const replicateSession = new ReplicateSession({
   replicateKey,
@@ -48,7 +53,13 @@ const results = await queryEngine.query({
 ## Full Example
 
 ```ts
-import { LlamaDeuce, Document, VectorStoreIndex, Settings } from "llamaindex";
+import {
+  LlamaDeuce,
+  Document,
+  VectorStoreIndex,
+  Settings,
+  DeuceChatStrategy,
+} from "llamaindex";
 
 // Use the LlamaDeuce LLM
 Settings.llm = new LlamaDeuce({ chatStrategy: DeuceChatStrategy.META });

apps/docs/docs/modules/node_postprocessors/jinaai_reranker.md

@@ -0,0 +1,71 @@
+# Jina AI Reranker
+
+The Jina AI Reranker is a postprocessor that uses the Jina AI Reranker API to rerank the results of a search query.
+
+## Setup
+
+Firstly, you will need to install the `llamaindex` package.
+
+```bash
+pnpm install llamaindex
+```
+
+Now, you will need to sign up for an API key at [Jina AI](https://jina.ai/reranker). Once you have your API key you can import the necessary modules and create a new instance of the `JinaAIReranker` class.
+
+```ts
+import {
+  JinaAIReranker,
+  Document,
+  OpenAI,
+  VectorStoreIndex,
+  Settings,
+} from "llamaindex";
+```
+
+## Load and index documents
+
+For this example, we will use a single document. In a real-world scenario, you would have multiple documents to index.
+
+```ts
+const document = new Document({ text: essay, id_: "essay" });
+
+Settings.llm = new OpenAI({ model: "gpt-3.5-turbo", temperature: 0.1 });
+
+const index = await VectorStoreIndex.fromDocuments([document]);
+```
+
+## Increase similarity topK to retrieve more results
+
+The default value for `similarityTopK` is 2. This means that only the most similar document will be returned. To retrieve more results, you can increase the value of `similarityTopK`.
+
+```ts
+const retriever = index.asRetriever();
+retriever.similarityTopK = 5;
+```
+
+## Create a new instance of the JinaAIReranker class
+
+Then you can create a new instance of the `JinaAIReranker` class and pass in the number of results you want to return.
+The Jina AI Reranker API key is set in the `JINAAI_API_KEY` environment variable.
+
+```bash
+export JINAAI_API_KEY=<YOUR API KEY>
+```
+
+```ts
+const nodePostprocessor = new JinaAIReranker({
+  topN: 5,
+});
+```
+
+## Create a query engine with the retriever and node postprocessor
+
+```ts
+const queryEngine = index.asQueryEngine({
+  retriever,
+  nodePostprocessors: [nodePostprocessor],
+});
+
+// log the response
+const response = await queryEngine.query("Where did the author grown up?");
+```

apps/docs/docs/observability/index.md

@@ -14,6 +14,9 @@ Configure a variable once, and you'll be able to do things like the following:
 
 Each provider has similarities and differences. Take a look below for the full set of guides for each one!
 
+- [OpenLLMetry](#openllmetry)
+- [Langtrace](#langtrace)
+
 ## OpenLLMetry
 
 [OpenLLMetry](https://github.com/traceloop/openllmetry-js) is an open-source project based on OpenTelemetry for tracing and monitoring
@@ -33,3 +36,29 @@ traceloop.initialize({
   disableBatch: true,
 });
 ```
+
+## Langtrace
+
+Enhance your observability with Langtrace, a robust open-source tool supports OpenTelemetry and is designed to trace, evaluate, and manage LLM applications seamlessly. Langtrace integrates directly with LlamaIndex, offering detailed, real-time insights into performance metrics such as accuracy, evaluations, and latency.
+
+#### Install
+
+- Self-host or sign-up and generate an API key using [Langtrace](https://www.langtrace.ai) Cloud
+
+```bash
+npm install @langtrase/typescript-sdk
+```
+
+#### Initialize
+
+```js
+import * as Langtrace from "@langtrase/typescript-sdk";
+Langtrace.init({ api_key: "<YOUR_API_KEY>" });
+```
+
+Features:
+
+- OpenTelemetry compliant, ensuring broad compatibility with observability platforms.
+- Provides comprehensive logs and detailed traces of all components.
+- Real-time monitoring of accuracy, evaluations, usage, costs, and latency.
+- For more configuration options and details, visit [Langtrace Docs](https://docs.langtrace.ai/introduction).

apps/docs/docs/recipes/cost-analysis.mdx

@@ -6,7 +6,7 @@ This page shows how to track LLM cost using APIs.
 
 The callback manager is a class that manages the callback functions.
 
-You can register `llm-start`, and `llm-end` callbacks to the callback manager for tracking the cost.
+You can register `llm-start`, `llm-end`, and `llm-stream` callbacks to the callback manager for tracking the cost.
 
 import CodeBlock from "@theme/CodeBlock";
 import CodeSource from "!raw-loader!../../../../examples/recipes/cost-analysis";

apps/docs/docusaurus.config.js

@@ -66,7 +66,11 @@ const config = {
             [require("@docusaurus/remark-plugin-npm2yarn"), { sync: true }],
           ],
         },
-        blog: false,
+        blog: {
+          blogTitle: "LlamaIndexTS blog",
+          blogDescription: "The official blog of LlamaIndexTS",
+          postsPerPage: "ALL",
+        },
         gtag: {
           trackingID: "G-NB9B8LW9W5",
           anonymizeIP: true,
@@ -97,6 +101,7 @@ const config = {
             type: "localeDropdown",
             position: "left",
           },
+          { to: "blog", label: "Blog", position: "right" },
           {
             href: "https://github.com/run-llama/LlamaIndexTS",
             label: "GitHub",
@@ -163,7 +168,7 @@ const config = {
       "docusaurus-plugin-typedoc",
       {
         entryPoints: ["../../packages/core/src/index.ts"],
-        tsconfig: "../../packages/core/tsconfig.json",
+        tsconfig: "../../tsconfig.json",
         readme: "none",
         sourceLinkTemplate:
           "https://github.com/run-llama/LlamaIndexTS/blob/{gitRevision}/{path}#L{line}",

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.4",
+  "version": "0.0.19",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",
@@ -15,28 +15,29 @@
     "typecheck": "tsc"
   },
   "dependencies": {
-    "@docusaurus/core": "^3.2.0",
-    "@docusaurus/remark-plugin-npm2yarn": "^3.2.0",
+    "@docusaurus/core": "^3.3.2",
+    "@docusaurus/remark-plugin-npm2yarn": "^3.3.2",
     "@llamaindex/examples": "workspace:*",
-    "@mdx-js/react": "^3.0.0",
-    "clsx": "^2.1.0",
-    "postcss": "^8.4.33",
+    "@mdx-js/react": "^3.0.1",
+    "clsx": "^2.1.1",
+    "llamaindex": "workspace:*",
+    "postcss": "^8.4.38",
     "prism-react-renderer": "^2.3.1",
     "raw-loader": "^4.0.2",
-    "react": "^18.2.0",
-    "react-dom": "^18.2.0"
+    "react": "^18.3.1",
+    "react-dom": "^18.3.1"
   },
   "devDependencies": {
-    "@docusaurus/module-type-aliases": "3.2.0",
-    "@docusaurus/preset-classic": "^3.2.0",
-    "@docusaurus/theme-classic": "^3.2.0",
-    "@docusaurus/types": "^3.2.0",
+    "@docusaurus/module-type-aliases": "3.3.2",
+    "@docusaurus/preset-classic": "^3.3.2",
+    "@docusaurus/theme-classic": "^3.3.2",
+    "@docusaurus/types": "^3.3.2",
     "@tsconfig/docusaurus": "^2.0.3",
-    "@types/node": "^18.19.10",
-    "docusaurus-plugin-typedoc": "^0.22.0",
-    "typedoc": "^0.25.12",
-    "typedoc-plugin-markdown": "^3.17.1",
-    "typescript": "^5.4.3"
+    "@types/node": "^20.12.11",
+    "docusaurus-plugin-typedoc": "^1.0.1",
+    "typedoc": "^0.25.13",
+    "typedoc-plugin-markdown": "^4.0.1",
+    "typescript": "^5.4.5"
   },
   "browserslist": {
     "production": [

examples/.env.example

@@ -0,0 +1 @@
+DEBUG=llamaindex

examples/agent/multi_document_agent.ts

@@ -86,7 +86,6 @@ async function main() {
     const agent = new OpenAIAgent({
       tools: queryEngineTools,
       llm: new OpenAI({ model: "gpt-4" }),
-      verbose: true,
     });
 
     documentAgents[title] = agent;
@@ -126,8 +125,7 @@ async function main() {
   const topAgent = new OpenAIAgent({
     toolRetriever: await objectIndex.asRetriever({}),
     llm: new OpenAI({ model: "gpt-4" }),
-    verbose: true,
-    prefixMessages: [
+    chatHistory: [
       {
         content:
           "You are an agent designed to answer queries about a set of given countries. Please always use the tools provided to answer a question. Do not rely on prior knowledge.",
@@ -145,4 +143,4 @@ async function main() {
   });
 }
 
-main();
+void main();

examples/agent/openai-task.ts

@@ -0,0 +1,39 @@
+import { ChatResponseChunk, OpenAIAgent } from "llamaindex";
+import { ReadableStream } from "node:stream/web";
+import {
+  getCurrentIDTool,
+  getUserInfoTool,
+  getWeatherTool,
+} from "./utils/tools";
+
+async function main() {
+  // Create an OpenAIAgent with the function tools
+  const agent = new OpenAIAgent({
+    tools: [getCurrentIDTool, getUserInfoTool, getWeatherTool],
+  });
+
+  const task = await agent.createTask(
+    "What is my current address weather based on my profile?",
+    true,
+  );
+
+  for await (const stepOutput of task) {
+    const stream = stepOutput.output as ReadableStream<ChatResponseChunk>;
+    if (stepOutput.isLast) {
+      for await (const chunk of stream) {
+        process.stdout.write(chunk.delta);
+      }
+      process.stdout.write("\n");
+    } else {
+      // handing function call
+      console.log("handling function call...");
+      for await (const chunk of stream) {
+        console.log("debug:", JSON.stringify(chunk.raw));
+      }
+    }
+  }
+}
+
+void main().then(() => {
+  console.log("Done");
+});

examples/agent/openai.ts

@@ -1,76 +1,61 @@
 import { FunctionTool, OpenAIAgent } from "llamaindex";
 
-// Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-
-// Define the parameters of the sum function as a JSON schema
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The dividend a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The divisor b to divide by",
-    },
-  },
-  required: ["a", "b"],
-};
-
-async function main() {
-  // Create a function tool from the sum function
-  const functionTool = new FunctionTool(sumNumbers, {
+const sumNumbers = FunctionTool.from(
+  ({ a, b }: { a: number; b: number }) => `${a + b}`,
+  {
     name: "sumNumbers",
     description: "Use this function to sum two numbers",
-    parameters: sumJSON,
-  });
+    parameters: {
+      type: "object",
+      properties: {
+        a: {
+          type: "number",
+          description: "The first number",
+        },
+        b: {
+          type: "number",
+          description: "The second number",
+        },
+      },
+      required: ["a", "b"],
+    },
+  },
+);
 
-  // Create a function tool from the divide function
-  const functionTool2 = new FunctionTool(divideNumbers, {
+const divideNumbers = FunctionTool.from(
+  ({ a, b }: { a: number; b: number }) => `${a / b}`,
+  {
     name: "divideNumbers",
     description: "Use this function to divide two numbers",
-    parameters: divideJSON,
-  });
+    parameters: {
+      type: "object",
+      properties: {
+        a: {
+          type: "number",
+          description: "The dividend a to divide",
+        },
+        b: {
+          type: "number",
+          description: "The divisor b to divide by",
+        },
+      },
+      required: ["a", "b"],
+    },
+  },
+);
 
-  // Create an OpenAIAgent with the function tools
+async function main() {
   const agent = new OpenAIAgent({
-    tools: [functionTool, functionTool2],
-    verbose: true,
+    tools: [sumNumbers, divideNumbers],
   });
 
-  // Chat with the agent
   const response = await agent.chat({
     message: "How much is 5 + 5? then divide by 2",
   });
 
-  // Print the response
-  console.log(String(response));
+  console.log(response.response.message);
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("Done");
 });

examples/agent/query_openai_agent.ts

@@ -34,13 +34,13 @@ async function main() {
 
   // Chat with the agent
   const response = await agent.chat({
-    message: "What was his salary?",
+    message: "What was his first salary?",
   });
 
   // Print the response
-  console.log(String(response));
+  console.log(response.response);
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("Done");
 });

examples/agent/react-task.ts

@@ -0,0 +1,40 @@
+import { ChatResponseChunk, ReActAgent } from "llamaindex";
+import { ReadableStream } from "node:stream/web";
+import {
+  getCurrentIDTool,
+  getUserInfoTool,
+  getWeatherTool,
+} from "./utils/tools";
+
+async function main() {
+  // Create an OpenAIAgent with the function tools
+  const agent = new ReActAgent({
+    tools: [getCurrentIDTool, getUserInfoTool, getWeatherTool],
+  });
+
+  const task = await agent.createTask(
+    "What is my current address weather based on my profile?",
+    true,
+  );
+
+  for await (const stepOutput of task) {
+    const stream = stepOutput.output as ReadableStream<ChatResponseChunk>;
+    if (stepOutput.isLast) {
+      for await (const chunk of stream) {
+        process.stdout.write(chunk.delta);
+      }
+      process.stdout.write("\n");
+    } else {
+      // handing function call
+      console.log("handling function call...");
+      for await (const chunk of stream) {
+        console.log("debug:", JSON.stringify(chunk.raw));
+      }
+    }
+    console.log("---");
+  }
+}
+
+void main().then(() => {
+  console.log("Done");
+});

examples/agent/react_agent.ts

@@ -1,13 +1,13 @@
 import { Anthropic, FunctionTool, ReActAgent } from "llamaindex";
 
 // Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
+function sumNumbers({ a, b }: { a: number; b: number }) {
+  return `${a + b}`;
 }
 
 // Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
+function divideNumbers({ a, b }: { a: number; b: number }) {
+  return `${a / b}`;
 }
 
 // Define the parameters of the sum function as a JSON schema
@@ -24,7 +24,7 @@ const sumJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 const divideJSON = {
   type: "object",
@@ -39,7 +39,7 @@ const divideJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 async function main() {
   // Create a function tool from the sum function
@@ -65,18 +65,17 @@ async function main() {
   const agent = new ReActAgent({
     llm: anthropic,
     tools: [functionTool, functionTool2],
-    verbose: true,
   });
 
   // Chat with the agent
-  const response = await agent.chat({
+  const { response } = await agent.chat({
     message: "Divide 16 by 2 then add 20",
   });
 
   // Print the response
-  console.log(String(response));
+  console.log(response.message);
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("Done");
 });

examples/agent/retriever_openai_agent.ts

@@ -0,0 +1,65 @@
+import {
+  FunctionTool,
+  MetadataMode,
+  NodeWithScore,
+  OpenAIAgent,
+  SimpleDirectoryReader,
+  VectorStoreIndex,
+} from "llamaindex";
+
+async function main() {
+  // Load the documents
+  const documents = await new SimpleDirectoryReader().loadData({
+    directoryPath: "node_modules/llamaindex/examples",
+  });
+
+  // Create a vector index from the documents
+  const vectorIndex = await VectorStoreIndex.fromDocuments(documents);
+
+  const retriever = vectorIndex.asRetriever({ similarityTopK: 3 });
+
+  const retrieverTool = FunctionTool.from(
+    async ({ query }: { query: string }) => {
+      const nodesWithScores = await retriever.retrieve({
+        query,
+      });
+      return nodesWithScores
+        .map((nodeWithScore: NodeWithScore) =>
+          nodeWithScore.node.getContent(MetadataMode.NONE),
+        )
+        .join("\n");
+    },
+    {
+      name: "get_abramov_info",
+      description: "Get information about the Abramov documents",
+      parameters: {
+        type: "object",
+        properties: {
+          query: {
+            type: "string",
+            description: "The query about Abramov",
+          },
+        },
+        required: ["query"],
+      },
+    },
+  );
+
+  // Create an OpenAIAgent with the function tools
+  const agent = new OpenAIAgent({
+    tools: [retrieverTool],
+    verbose: true,
+  });
+
+  // Chat with the agent
+  const response = await agent.chat({
+    message: "What was his first salary?",
+  });
+
+  // Print the response
+  console.log(response.response);
+}
+
+void main().then(() => {
+  console.log("Done");
+});

examples/agent/step_wise_openai.ts

@@ -1,95 +0,0 @@
-import { FunctionTool, OpenAIAgent } from "llamaindex";
-
-// Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
-}
-
-// Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
-}
-
-// Define the parameters of the sum function as a JSON schema
-const sumJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The first number",
-    },
-    b: {
-      type: "number",
-      description: "The second number",
-    },
-  },
-  required: ["a", "b"],
-};
-
-const divideJSON = {
-  type: "object",
-  properties: {
-    a: {
-      type: "number",
-      description: "The dividend a to divide",
-    },
-    b: {
-      type: "number",
-      description: "The divisor b to divide by",
-    },
-  },
-  required: ["a", "b"],
-};
-
-async function main() {
-  // Create a function tool from the sum function
-  const functionTool = new FunctionTool(sumNumbers, {
-    name: "sumNumbers",
-    description: "Use this function to sum two numbers",
-    parameters: sumJSON,
-  });
-
-  // Create a function tool from the divide function
-  const functionTool2 = new FunctionTool(divideNumbers, {
-    name: "divideNumbers",
-    description: "Use this function to divide two numbers",
-    parameters: divideJSON,
-  });
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [functionTool, functionTool2],
-    verbose: true,
-  });
-
-  // Create a task to sum and divide numbers
-  const task = agent.createTask("How much is 5 + 5? then divide by 2");
-
-  let count = 0;
-
-  while (true) {
-    const stepOutput = await agent.runStep(task.taskId);
-
-    console.log(`Runnning step ${count++}`);
-    console.log(`======== OUTPUT ==========`);
-    if (stepOutput.output.response) {
-      console.log(stepOutput.output.response);
-    } else {
-      console.log(stepOutput.output.sources);
-    }
-    console.log(`==========================`);
-
-    if (stepOutput.isLast) {
-      const finalResponse = await agent.finalizeResponse(
-        task.taskId,
-        stepOutput,
-      );
-      console.log({ finalResponse });
-      break;
-    }
-  }
-}
-
-main().then(() => {
-  console.log("Done");
-});

examples/agent/step_wise_query_tool.ts

@@ -29,36 +29,15 @@ async function main() {
   // Create an OpenAIAgent with the function tools
   const agent = new OpenAIAgent({
     tools: [queryEngineTool],
-    verbose: true,
   });
 
-  const task = agent.createTask("What was his salary?");
+  const { response } = await agent.chat({
+    message: "What was his salary?",
+  });
 
-  let count = 0;
-
-  while (true) {
-    const stepOutput = await agent.runStep(task.taskId);
-
-    console.log(`Runnning step ${count++}`);
-    console.log(`======== OUTPUT ==========`);
-    if (stepOutput.output.response) {
-      console.log(stepOutput.output.response);
-    } else {
-      console.log(stepOutput.output.sources);
-    }
-    console.log(`==========================`);
-
-    if (stepOutput.isLast) {
-      const finalResponse = await agent.finalizeResponse(
-        task.taskId,
-        stepOutput,
-      );
-      console.log({ finalResponse });
-      break;
-    }
-  }
+  console.log(response.message.content);
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("Done");
 });

examples/agent/step_wise_react.ts

@@ -1,13 +1,14 @@
-import { FunctionTool, ReActAgent } from "llamaindex";
+import { Anthropic, FunctionTool, ReActAgent } from "llamaindex";
 
 // Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
+function sumNumbers({ a, b }: { a: number; b: number }) {
+  return `${a + b}`;
 }
 
 // Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
+function divideNumbers({ a, b }: { a: number; b: number }) {
+  console.log("get input", a, b);
+  return `${a / b}`;
 }
 
 // Define the parameters of the sum function as a JSON schema
@@ -24,7 +25,7 @@ const sumJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 const divideJSON = {
   type: "object",
@@ -39,7 +40,7 @@ const divideJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 async function main() {
   // Create a function tool from the sum function
@@ -58,33 +59,24 @@ async function main() {
 
   // Create an OpenAIAgent with the function tools
   const agent = new ReActAgent({
+    llm: new Anthropic({
+      model: "claude-3-opus",
+    }),
     tools: [functionTool, functionTool2],
-    verbose: true,
   });
 
-  const task = agent.createTask("Divide 16 by 2 then add 20");
+  const task = await agent.createTask("Divide 16 by 2 then add 20");
 
   let count = 0;
 
-  while (true) {
-    const stepOutput = await agent.runStep(task.taskId);
-
+  for await (const stepOutput of task) {
     console.log(`Runnning step ${count++}`);
     console.log(`======== OUTPUT ==========`);
-    console.log(stepOutput.output);
+    console.log(stepOutput);
     console.log(`==========================`);
-
-    if (stepOutput.isLast) {
-      const finalResponse = await agent.finalizeResponse(
-        task.taskId,
-        stepOutput,
-      );
-      console.log({ finalResponse });
-      break;
-    }
   }
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("Done");
 });

examples/agent/stream_openai_agent.ts

@@ -1,13 +1,13 @@
 import { FunctionTool, OpenAIAgent } from "llamaindex";
 
 // Define a function to sum two numbers
-function sumNumbers({ a, b }: { a: number; b: number }): number {
-  return a + b;
+function sumNumbers({ a, b }: { a: number; b: number }) {
+  return `${a + b}`;
 }
 
 // Define a function to divide two numbers
-function divideNumbers({ a, b }: { a: number; b: number }): number {
-  return a / b;
+function divideNumbers({ a, b }: { a: number; b: number }) {
+  return `${a / b}`;
 }
 
 // Define the parameters of the sum function as a JSON schema
@@ -24,7 +24,7 @@ const sumJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 const divideJSON = {
   type: "object",
@@ -39,18 +39,18 @@ const divideJSON = {
     },
   },
   required: ["a", "b"],
-};
+} as const;
 
 async function main() {
   // Create a function tool from the sum function
-  const functionTool = new FunctionTool(sumNumbers, {
+  const functionTool = FunctionTool.from(sumNumbers, {
     name: "sumNumbers",
     description: "Use this function to sum two numbers",
     parameters: sumJSON,
   });
 
   // Create a function tool from the divide function
-  const functionTool2 = new FunctionTool(divideNumbers, {
+  const functionTool2 = FunctionTool.from(divideNumbers, {
     name: "divideNumbers",
     description: "Use this function to divide two numbers",
     parameters: divideJSON,
@@ -59,7 +59,6 @@ async function main() {
   // Create an OpenAIAgent with the function tools
   const agent = new OpenAIAgent({
     tools: [functionTool, functionTool2],
-    verbose: false,
   });
 
   const stream = await agent.chat({
@@ -67,11 +66,15 @@ async function main() {
     stream: true,
   });
 
-  for await (const chunk of stream.response) {
-    process.stdout.write(chunk.response);
+  console.log("Response:");
+
+  for await (const {
+    response: { delta },
+  } of stream) {
+    process.stdout.write(delta);
   }
 }
 
-main().then(() => {
+void main().then(() => {
   console.log("\nDone");
 });

examples/agent/utils/tools.ts

@@ -0,0 +1,54 @@
+import { FunctionTool } from "llamaindex";
+
+export const getCurrentIDTool = FunctionTool.from(
+  () => {
+    console.log("Getting user id...");
+    return crypto.randomUUID();
+  },
+  {
+    name: "get_user_id",
+    description: "Get a random user id",
+  },
+);
+
+export const getUserInfoTool = FunctionTool.from(
+  ({ userId }: { userId: string }) => {
+    console.log("Getting user info...", userId);
+    return `Name: Alex; Address: 1234 Main St, CA; User ID: ${userId}`;
+  },
+  {
+    name: "get_user_info",
+    description: "Get user info",
+    parameters: {
+      type: "object",
+      properties: {
+        userId: {
+          type: "string",
+          description: "The user id",
+        },
+      },
+      required: ["userId"],
+    },
+  },
+);
+
+export const getWeatherTool = FunctionTool.from(
+  ({ address }: { address: string }) => {
+    console.log("Getting weather...", address);
+    return `${address} is in a sunny location!`;
+  },
+  {
+    name: "get_weather",
+    description: "Get the current weather for a location",
+    parameters: {
+      type: "object",
+      properties: {
+        address: {
+          type: "string",
+          description: "The address",
+        },
+      },
+      required: ["address"],
+    },
+  },
+);

examples/agent/wiki.ts

@@ -0,0 +1,29 @@
+import { OpenAI, OpenAIAgent, WikipediaTool } from "llamaindex";
+
+async function main() {
+  const llm = new OpenAI({ model: "gpt-4-turbo" });
+  const wikiTool = new WikipediaTool();
+
+  // Create an OpenAIAgent with the Wikipedia tool
+  const agent = new OpenAIAgent({
+    llm,
+    tools: [wikiTool],
+  });
+
+  // Chat with the agent
+  const response = await agent.chat({
+    message: "Who was Goethe?",
+    stream: true,
+  });
+
+  for await (const {
+    response: { delta },
+  } of response) {
+    process.stdout.write(delta);
+  }
+}
+
+(async function () {
+  await main();
+  console.log("\nDone");
+})();

examples/agent/wikipedia-tool.ts

@@ -1,23 +0,0 @@
-import { OpenAIAgent, WikipediaTool } from "llamaindex";
-
-async function main() {
-  const wikipediaTool = new WikipediaTool();
-
-  // Create an OpenAIAgent with the function tools
-  const agent = new OpenAIAgent({
-    tools: [wikipediaTool],
-    verbose: true,
-  });
-
-  // Chat with the agent
-  const response = await agent.chat({
-    message: "Where is Ho Chi Minh City?",
-  });
-
-  // Print the response
-  console.log(response);
-}
-
-main().then(() => {
-  console.log("Done");
-});

examples/anthropic/agent.ts

@@ -0,0 +1,43 @@
+import { FunctionTool, Settings, WikipediaTool } from "llamaindex";
+import { AnthropicAgent } from "llamaindex/agent/anthropic";
+
+Settings.callbackManager.on("llm-tool-call", (event) => {
+  console.log("llm-tool-call", event.detail.payload.toolCall);
+});
+
+const agent = new AnthropicAgent({
+  tools: [
+    FunctionTool.from<{ location: string }>(
+      (query) => {
+        return `The weather in ${query.location} is sunny`;
+      },
+      {
+        name: "weather",
+        description: "Get the weather",
+        parameters: {
+          type: "object",
+          properties: {
+            location: {
+              type: "string",
+              description: "The location to get the weather for",
+            },
+          },
+          required: ["location"],
+        },
+      },
+    ),
+    new WikipediaTool(),
+  ],
+});
+
+async function main() {
+  // https://docs.anthropic.com/claude/docs/tool-use#tool-use-best-practices-and-limitations
+  const { response } = await agent.chat({
+    message:
+      "What is the weather in New York? What's the history of New York from Wikipedia in 3 sentences?",
+  });
+
+  console.log(response);
+}
+
+void main();

examples/astradb/README.md

@@ -13,7 +13,7 @@ Here are two sample scripts which work well with the sample data in the Astra Po
 1. Set your env variables:
 
 - `ASTRA_DB_APPLICATION_TOKEN`: The generated app token for your Astra database
-- `ASTRA_DB_ENDPOINT`: The API endpoint for your Astra database
+- `ASTRA_DB_API_ENDPOINT`: The API endpoint for your Astra database
 - `ASTRA_DB_NAMESPACE`: (Optional) The namespace where your collection is stored defaults to `default_keyspace`
 - `OPENAI_API_KEY`: Your OpenAI key
 

examples/astradb/example.ts

@@ -34,10 +34,9 @@ async function main() {
     ];
 
     const astraVS = new AstraDBVectorStore();
-    await astraVS.create(collectionName, {
+    await astraVS.createAndConnect(collectionName, {
       vector: { dimension: 1536, metric: "cosine" },
     });
-    await astraVS.connect(collectionName);
 
     const ctx = await storageContextFromDefaults({ vectorStore: astraVS });
     const index = await VectorStoreIndex.fromDocuments(docs, {
@@ -55,4 +54,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/astradb/load.ts

@@ -13,7 +13,7 @@ async function main() {
     const docs = await reader.loadData("./data/movie_reviews.csv");
 
     const astraVS = new AstraDBVectorStore({ contentKey: "reviewtext" });
-    await astraVS.create(collectionName, {
+    await astraVS.createAndConnect(collectionName, {
       vector: { dimension: 1536, metric: "cosine" },
     });
     await astraVS.connect(collectionName);
@@ -27,4 +27,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/astradb/query.ts

@@ -1,4 +1,8 @@
-import { AstraDBVectorStore, VectorStoreIndex } from "llamaindex";
+import {
+  AstraDBVectorStore,
+  VectorStoreIndex,
+  serviceContextFromDefaults,
+} from "llamaindex";
 
 const collectionName = "movie_reviews";
 
@@ -7,7 +11,8 @@ async function main() {
     const astraVS = new AstraDBVectorStore({ contentKey: "reviewtext" });
     await astraVS.connect(collectionName);
 
-    const index = await VectorStoreIndex.fromVectorStore(astraVS);
+    const ctx = serviceContextFromDefaults();
+    const index = await VectorStoreIndex.fromVectorStore(astraVS, ctx);
 
     const retriever = await index.asRetriever({ similarityTopK: 20 });
 
@@ -23,4 +28,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/chatHistory.ts

@@ -1,7 +1,18 @@
 import { stdin as input, stdout as output } from "node:process";
 import readline from "node:readline/promises";
 
-import { OpenAI, SimpleChatEngine, SummaryChatHistory } from "llamaindex";
+import {
+  OpenAI,
+  Settings,
+  SimpleChatEngine,
+  SummaryChatHistory,
+} from "llamaindex";
+
+if (process.env.NODE_ENV === "development") {
+  Settings.callbackManager.on("llm-end", (event) => {
+    console.log("callers chain", event.reason?.computedCallers);
+  });
+}
 
 async function main() {
   // Set maxTokens to 75% of the context window size of 4096

examples/chromadb/preFilters.ts

@@ -54,4 +54,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/chromadb/test.ts

@@ -37,4 +37,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/cloud/from_documents.ts

@@ -0,0 +1,44 @@
+import fs from "node:fs/promises";
+
+import { stdin as input, stdout as output } from "node:process";
+
+import readline from "node:readline/promises";
+
+import { Document, LlamaCloudIndex } from "llamaindex";
+
+async function main() {
+  const path = "node_modules/llamaindex/examples/abramov.txt";
+
+  const essay = await fs.readFile(path, "utf-8");
+
+  // Create Document object with essay
+  const document = new Document({ text: essay, id_: path });
+
+  const index = await LlamaCloudIndex.fromDocuments({
+    documents: [document],
+    name: "test",
+    projectName: "default",
+    apiKey: process.env.LLAMA_CLOUD_API_KEY,
+    baseUrl: process.env.LLAMA_CLOUD_BASE_URL,
+  });
+
+  const queryEngine = index.asQueryEngine({
+    denseSimilarityTopK: 5,
+  });
+
+  const rl = readline.createInterface({ input, output });
+
+  while (true) {
+    const query = await rl.question("Query: ");
+    const stream = await queryEngine.query({
+      query,
+      stream: true,
+    });
+    console.log();
+    for await (const chunk of stream) {
+      process.stdout.write(chunk.response);
+    }
+  }
+}
+
+main().catch(console.error);

examples/evaluation/correctness.ts

@@ -22,4 +22,4 @@ However, general relativity, published in 1915, extended these ideas to include
   console.log(result);
 }
 
-main();
+void main();

examples/evaluation/faithfulness.ts

@@ -36,4 +36,4 @@ async function main() {
   console.log(result);
 }
 
-main();
+void main();

examples/evaluation/relevancy.ts

@@ -37,4 +37,4 @@ async function main() {
   console.log(result);
 }
 
-main();
+void main();

examples/gemini/chat.ts

@@ -0,0 +1,21 @@
+import { Gemini, GEMINI_MODEL } from "llamaindex";
+
+(async () => {
+  if (!process.env.GOOGLE_API_KEY) {
+    throw new Error("Please set the GOOGLE_API_KEY environment variable.");
+  }
+  const gemini = new Gemini({
+    model: GEMINI_MODEL.GEMINI_PRO,
+  });
+  const result = await gemini.chat({
+    messages: [
+      { content: "You want to talk in rhymes.", role: "system" },
+      {
+        content:
+          "How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
+        role: "user",
+      },
+    ],
+  });
+  console.log(result);
+})();

examples/gemini/embedding.ts

@@ -0,0 +1,15 @@
+import { GEMINI_EMBEDDING_MODEL, GeminiEmbedding } from "llamaindex";
+
+async function main() {
+  if (!process.env.GOOGLE_API_KEY) {
+    throw new Error("Please set the GOOGLE_API_KEY environment variable.");
+  }
+  const embedModel = new GeminiEmbedding({
+    model: GEMINI_EMBEDDING_MODEL.EMBEDDING_001,
+  });
+  const texts = ["hello", "world"];
+  const embeddings = await embedModel.getTextEmbeddingsBatch(texts);
+  console.log(`\nWe have ${embeddings.length} embeddings`);
+}
+
+main().catch(console.error);

examples/gptturbollama3.ts

@@ -0,0 +1,40 @@
+import { stdin as input, stdout as output } from "node:process";
+import readline from "node:readline/promises";
+
+import { ChatMessage, OpenAI, ReplicateLLM } from "llamaindex";
+
+(async () => {
+  const gpt4 = new OpenAI({ model: "gpt-4-turbo", temperature: 0.9 });
+  const l3 = new ReplicateLLM({
+    model: "llama-3-70b-instruct",
+    temperature: 0.9,
+  });
+
+  const rl = readline.createInterface({ input, output });
+  const start = await rl.question("Start: ");
+  const history: ChatMessage[] = [
+    {
+      content:
+        "Prefer shorter answers. Keep your response to 100 words or less.",
+      role: "system",
+    },
+    { content: start, role: "user" },
+  ];
+
+  while (true) {
+    const next = history.length % 2 === 1 ? gpt4 : l3;
+    const r = await next.chat({
+      messages: history.map(({ content, role }) => ({
+        content,
+        role: next === l3 ? role : role === "user" ? "assistant" : "user",
+      })),
+    });
+    history.push({
+      content: r.message.content,
+      role: next === l3 ? "assistant" : "user",
+    });
+    await rl.question(
+      (next === l3 ? "Llama 3: " : "GPT 4 Turbo: ") + r.message.content,
+    );
+  }
+})();

examples/huggingface/chat.ts

@@ -0,0 +1,22 @@
+import { HuggingFaceInferenceAPI } from "llamaindex";
+
+(async () => {
+  if (!process.env.HUGGING_FACE_TOKEN) {
+    throw new Error("Please set the HUGGING_FACE_TOKEN environment variable.");
+  }
+  const hf = new HuggingFaceInferenceAPI({
+    accessToken: process.env.HUGGING_FACE_TOKEN,
+    model: "mistralai/Mixtral-8x7B-Instruct-v0.1",
+  });
+  const result = await hf.chat({
+    messages: [
+      { content: "You want to talk in rhymes.", role: "system" },
+      {
+        content:
+          "How much wood would a woodchuck chuck if a woodchuck could chuck wood?",
+        role: "user",
+      },
+    ],
+  });
+  console.log(result);
+})();

examples/jsonExtract.ts

@@ -36,9 +36,7 @@ async function main() {
     ],
   });
 
-  const json = JSON.parse(response.message.content);
-
-  console.log(json);
+  console.log(response.message.content);
 }
 
 main().catch(console.error);

examples/llama3.ts

@@ -0,0 +1,13 @@
+import { ReplicateLLM } from "llamaindex";
+
+(async () => {
+  const tres = new ReplicateLLM({ model: "llama-3-70b-instruct" });
+  const stream = await tres.chat({
+    messages: [{ content: "Hello, world!", role: "user" }],
+    stream: true,
+  });
+  for await (const chunk of stream) {
+    process.stdout.write(chunk.delta);
+  }
+  console.log("\n\ndone");
+})();

examples/milvus/load.ts

@@ -23,4 +23,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/milvus/query.ts

@@ -22,4 +22,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/mongo.ts

@@ -61,4 +61,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/mongodb/1_import.ts

@@ -31,4 +31,4 @@ async function importJsonToMongo() {
 }
 
 // Run the import function
-importJsonToMongo();
+void importJsonToMongo();

examples/mongodb/3_query.ts

@@ -27,4 +27,4 @@ async function query() {
   await client.close();
 }
 
-query();
+void query();

examples/multimodal/clip.ts

@@ -30,4 +30,4 @@ async function main() {
   console.log(`Similarity between "${text2}" and the image is ${sim2}`);
 }
 
-main();
+void main();

examples/multimodal/rag.ts

@@ -13,7 +13,7 @@ Settings.chunkSize = 512;
 Settings.chunkOverlap = 20;
 
 // Update llm
-Settings.llm = new OpenAI({ model: "gpt-4-vision-preview", maxTokens: 512 });
+Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 
 // Update callbackManager
 Settings.callbackManager = new CallbackManager({

examples/nodeParser/MarkdownNodeParser.ts

@@ -21,4 +21,4 @@ Sub-header content
   console.log(splits);
 }
 
-main();
+void main();

examples/ollama.ts

@@ -1,7 +1,14 @@
+import { OllamaEmbedding } from "llamaindex";
 import { Ollama } from "llamaindex/llm/ollama";
 
 (async () => {
-  const llm = new Ollama({ model: "llama2", temperature: 0.75 });
+  const llm = new Ollama({
+    model: "llama3",
+    config: {
+      host: "http://localhost:11434",
+    },
+  });
+  const embedModel = new OllamaEmbedding({ model: "nomic-embed-text" });
   {
     const response = await llm.chat({
       messages: [{ content: "Tell me a joke.", role: "user" }],
@@ -35,7 +42,7 @@ import { Ollama } from "llamaindex/llm/ollama";
     console.log(); // newline
   }
   {
-    const embedding = await llm.getTextEmbedding("Hello world!");
+    const embedding = await embedModel.getTextEmbedding("Hello world!");
     console.log("Embedding:", embedding);
   }
 })();

examples/package.json

@@ -4,24 +4,28 @@
   "version": "0.0.4",
   "dependencies": {
     "@aws-crypto/sha256-js": "^5.2.0",
-    "@datastax/astra-db-ts": "^0.1.4",
-    "@notionhq/client": "^2.2.14",
-    "@pinecone-database/pinecone": "^1.1.3",
-    "@zilliz/milvus2-sdk-node": "^2.3.5",
-    "chromadb": "^1.8.1",
-    "commander": "^11.1.0",
-    "dotenv": "^16.4.1",
-    "js-tiktoken": "^1.0.10",
-    "llamaindex": "latest",
-    "mongodb": "^6.2.0",
+    "@datastax/astra-db-ts": "^1.1.0",
+    "@notionhq/client": "^2.2.15",
+    "@pinecone-database/pinecone": "^2.2.0",
+    "@zilliz/milvus2-sdk-node": "^2.4.2",
+    "chromadb": "^1.7.3",
+    "commander": "^12.0.0",
+    "dotenv": "^16.4.5",
+    "js-tiktoken": "^1.0.11",
+    "llamaindex": "*",
+    "mongodb": "^6.6.1",
     "pathe": "^1.1.2"
   },
   "devDependencies": {
-    "@types/node": "^18.19.10",
+    "@types/node": "^20.12.11",
     "ts-node": "^10.9.2",
-    "typescript": "^5.4.3"
+    "tsx": "^4.9.3",
+    "typescript": "^5.4.5"
   },
   "scripts": {
     "lint": "eslint ."
+  },
+  "stackblitz": {
+    "startCommand": "npm start"
   }
 }

examples/pg-vector-store/load-docs.ts

@@ -32,7 +32,7 @@ async function main(args: any) {
   console.log(`Found ${count} files`);
 
   console.log(`Importing contents from ${count} files in ${sourceDir}`);
-  var fileName = "";
+  const fileName = "";
   try {
     // Passing callback fn to the ctor here
     // will enable looging to console.
@@ -42,7 +42,7 @@ async function main(args: any) {
 
     const pgvs = new PGVectorStore();
     pgvs.setCollection(sourceDir);
-    pgvs.clearCollection();
+    await pgvs.clearCollection();
 
     const ctx = await storageContextFromDefaults({ vectorStore: pgvs });
 
@@ -65,4 +65,4 @@ async function main(args: any) {
   process.exit(0);
 }
 
-main(process.argv).catch((err) => console.error(err));
+void main(process.argv).catch((err) => console.error(err));

examples/pinecone-vector-store/load-docs.ts

@@ -32,7 +32,7 @@ async function main(args: any) {
   console.log(`Found ${count} files`);
 
   console.log(`Importing contents from ${count} files in ${sourceDir}`);
-  var fileName = "";
+  const fileName = "";
   try {
     // Passing callback fn to the ctor here
     // will enable looging to console.
@@ -63,4 +63,4 @@ async function main(args: any) {
   process.exit(0);
 }
 
-main(process.argv).catch((err) => console.error(err));
+void main(process.argv).catch((err) => console.error(err));

examples/prompts/promptMixin.ts

@@ -45,4 +45,4 @@ async function main() {
   await queryEngine.query({ query });
 }
 
-main();
+void main();

examples/qdrantdb/preFilters.ts

@@ -79,4 +79,4 @@ async function main() {
   }
 }
 
-main();
+void main();

examples/readers/package.json

@@ -3,20 +3,22 @@
   "private": true,
   "type": "module",
   "scripts": {
-    "start": "node --loader ts-node/esm ./src/simple-directory-reader.ts",
-    "start:csv": "node --loader ts-node/esm ./src/csv.ts",
-    "start:docx": "node --loader ts-node/esm ./src/docx.ts",
-    "start:html": "node --loader ts-node/esm ./src/html.ts",
-    "start:markdown": "node --loader ts-node/esm ./src/markdown.ts",
-    "start:pdf": "node --loader ts-node/esm ./src/pdf.ts",
-    "start:llamaparse": "node --loader ts-node/esm ./src/llamaparse.ts"
+    "start": "node --import tsx ./src/simple-directory-reader.ts",
+    "start:csv": "node --import tsx ./src/csv.ts",
+    "start:docx": "node --import tsx ./src/docx.ts",
+    "start:html": "node --import tsx ./src/html.ts",
+    "start:markdown": "node --import tsx ./src/markdown.ts",
+    "start:pdf": "node --import tsx ./src/pdf.ts",
+    "start:llamaparse": "node --import tsx ./src/llamaparse.ts",
+    "start:notion": "node --import tsx ./src/notion.ts",
+    "start:llamaparse2": "node --import tsx ./src/llamaparse_2.ts"
   },
   "dependencies": {
-    "llamaindex": "latest"
+    "llamaindex": "*"
   },
   "devDependencies": {
-    "@types/node": "^20.11.14",
-    "ts-node": "^10.9.2",
-    "typescript": "^5.4.3"
+    "@types/node": "^20.12.11",
+    "tsx": "^4.9.3",
+    "typescript": "^5.4.5"
   }
 }

examples/readers/src/docx.ts

@@ -20,4 +20,4 @@ async function main() {
   console.log(`Test query > ${SAMPLE_QUERY}:\n`, response.toString());
 }
 
-main();
+void main();

examples/readers/src/llamaparse_2.ts

@@ -0,0 +1,26 @@
+import fs from "fs/promises";
+import { LlamaParseReader } from "llamaindex";
+
+async function main() {
+  // Load PDF using LlamaParse. set apiKey here or in environment variable LLAMA_CLOUD_API_KEY
+  const reader = new LlamaParseReader({
+    resultType: "markdown",
+    language: "en",
+    parsingInstruction:
+      "The provided document is a manga comic book. Most pages do NOT have title. It does not contain tables. Try to reconstruct the dialogue happening in a cohesive way. Output any math equation in LATEX markdown (between $$)",
+  });
+  const documents = await reader.loadData("../data/manga.pdf"); // The manga.pdf in the data folder is just a copy of the TOS, due to copyright laws. You have to place your own. I used "The Manga Guide to Calculus" by Hiroyuki Kojima
+
+  // Assuming documents contain an array of pages or sections
+  const parsedManga = documents.map((page) => page.text).join("\n---\n");
+
+  // Output the parsed manga to .md file. Will be placed in ../example/readers/
+  try {
+    await fs.writeFile("./parsedManga.md", parsedManga);
+    console.log("Output successfully written to parsedManga.md");
+  } catch (err) {
+    console.error("Error writing to file:", err);
+  }
+}
+
+main().catch(console.error);

examples/readers/src/markdown.ts

@@ -20,4 +20,4 @@ async function main() {
   console.log(`Test query > ${SAMPLE_QUERY}:\n`, response.toString());
 }
 
-main();
+void main();

examples/readers/src/notion.ts

@@ -7,7 +7,7 @@ import { createInterface } from "node:readline/promises";
 
 program
   .argument("[page]", "Notion page id (must be provided)")
-  .action(async (page, _options, command) => {
+  .action(async (page, _options) => {
     // Initializing a client
 
     if (!process.env.NOTION_TOKEN) {
@@ -55,7 +55,7 @@ program
           .filter((page) => page !== null);
         console.log("Found pages:");
         console.table(pages);
-        console.log(`To run, run ts-node ${command.name()} [page id]`);
+        console.log(`To run, run with [page id]`);
         return;
       }
     }

examples/recipes/cost-analysis.ts

@@ -1,37 +1,46 @@
 import { encodingForModel } from "js-tiktoken";
-import { OpenAI } from "llamaindex";
+import { ChatMessage, OpenAI, type LLMStartEvent } from "llamaindex";
 import { Settings } from "llamaindex/Settings";
+import { extractText } from "llamaindex/llm/utils";
 
 const encoding = encodingForModel("gpt-4-0125-preview");
 
 const llm = new OpenAI({
-  model: "gpt-4-0125-preview",
+  // currently is "gpt-4-turbo-2024-04-09"
+  model: "gpt-4-turbo",
 });
 
 let tokenCount = 0;
 
-Settings.callbackManager.on("llm-start", (event) => {
+Settings.callbackManager.on("llm-start", (event: LLMStartEvent) => {
   const { messages } = event.detail.payload;
-  tokenCount += messages.reduce((count, message) => {
-    return count + encoding.encode(message.content).length;
+  messages.reduce((count: number, message: ChatMessage) => {
+    return count + encoding.encode(extractText(message.content)).length;
   }, 0);
   console.log("Token count:", tokenCount);
   // https://openai.com/pricing
   // $10.00 / 1M tokens
-  console.log(`Price: $${(tokenCount / 1_000_000) * 10}`);
-});
-Settings.callbackManager.on("llm-end", (event) => {
-  const { response } = event.detail.payload;
-  tokenCount += encoding.encode(response.message.content).length;
-  console.log("Token count:", tokenCount);
-  // https://openai.com/pricing
-  // $30.00 / 1M tokens
-  console.log(`Price: $${(tokenCount / 1_000_000) * 30}`);
+  console.log(`Total Price: $${(tokenCount / 1_000_000) * 10}`);
 });
 
-const question = "Hello, how are you?";
+Settings.callbackManager.on("llm-stream", (event) => {
+  const { chunk } = event.detail.payload;
+  const { delta } = chunk;
+  tokenCount += encoding.encode(extractText(delta)).length;
+  if (tokenCount > 20) {
+    // This is just an example, you can set your own limit or handle it differently
+    throw new Error("Token limit exceeded!");
+  }
+});
+Settings.callbackManager.on("llm-end", () => {
+  // https://openai.com/pricing
+  // $30.00 / 1M tokens
+  console.log(`Total Price: $${(tokenCount / 1_000_000) * 30}`);
+});
+
+const question = "Hello, how are you? Please response about 50 tokens.";
 console.log("Question:", question);
-llm
+void llm
   .chat({
     stream: true,
     messages: [

examples/routerQueryEngine.ts

@@ -65,4 +65,4 @@ async function main() {
   });
 }
 
-main().then(() => console.log("Done"));
+void main().then(() => console.log("Done"));

examples/split.ts

@@ -13,4 +13,4 @@ async function main() {
   console.log(chunks);
 }
 
-main();
+void main();

examples/toolsStream.ts

@@ -0,0 +1,48 @@
+import { OpenAI } from "llamaindex";
+
+async function main() {
+  const llm = new OpenAI({ model: "gpt-4-turbo" });
+  const args: Parameters<typeof llm.chat>[0] = {
+    additionalChatOptions: {
+      tool_choice: "auto",
+    },
+    messages: [
+      {
+        content: "Who was Goethe?",
+        role: "user",
+      },
+    ],
+    tools: [
+      {
+        metadata: {
+          name: "wikipedia_tool",
+          description: "A tool that uses a query engine to search Wikipedia.",
+          parameters: {
+            type: "object",
+            properties: {
+              query: {
+                type: "string",
+                description: "The query to search for",
+              },
+            },
+            required: ["query"],
+          },
+        },
+      },
+    ],
+  };
+
+  const stream = await llm.chat({ ...args, stream: true });
+  for await (const chunk of stream) {
+    process.stdout.write(chunk.delta);
+    if (chunk.options && "toolCall" in chunk.options) {
+      console.log("Tool call:");
+      console.log(chunk.options.toolCall);
+    }
+  }
+}
+
+(async function () {
+  await main();
+  console.log("Done");
+})();

examples/tsconfig.json

@@ -1,12 +1,14 @@
 {
   "compilerOptions": {
-    "target": "es2017",
+    "target": "ES2022",
     "module": "esnext",
     "moduleResolution": "bundler",
     "esModuleInterop": true,
     "forceConsistentCasingInFileNames": true,
     "strict": true,
     "skipLibCheck": true,
+    "lib": ["ES2022"],
+    "types": ["node"],
     "outDir": "./lib",
     "tsBuildInfoFile": "./lib/.tsbuildinfo",
     "incremental": true,

examples/vectorIndex.ts

@@ -1,6 +1,11 @@
 import fs from "node:fs/promises";
 
-import { Document, VectorStoreIndex } from "llamaindex";
+import {
+  Document,
+  MetadataMode,
+  NodeWithScore,
+  VectorStoreIndex,
+} from "llamaindex";
 
 async function main() {
   // Load essay from abramov.txt in Node
@@ -16,12 +21,20 @@ async function main() {
 
   // Query the index
   const queryEngine = index.asQueryEngine();
-  const response = await queryEngine.query({
+  const { response, sourceNodes } = await queryEngine.query({
     query: "What did the author do in college?",
   });
 
-  // Output response
-  console.log(response.toString());
+  // Output response with sources
+  console.log(response);
+
+  if (sourceNodes) {
+    sourceNodes.forEach((source: NodeWithScore, index: number) => {
+      console.log(
+        `\n${index}: Score: ${source.score} - ${source.node.getContent(MetadataMode.NONE).substring(0, 50)}...\n`,
+      );
+    });
+  }
 }
 
 main().catch(console.error);

examples/vision.ts

@@ -1,7 +1,7 @@
 import { OpenAI } from "llamaindex";
 
 (async () => {
-  const llm = new OpenAI({ model: "gpt-4-vision-preview", temperature: 0.1 });
+  const llm = new OpenAI({ model: "gpt-4-turbo", temperature: 0.1 });
 
   // complete api
   const response1 = await llm.complete({ prompt: "How are you?" });

package.json

@@ -2,38 +2,44 @@
   "name": "@llamaindex/monorepo",
   "private": true,
   "scripts": {
-    "build": "turbo run build",
-    "build:release": "turbo run build lint test --filter=\"!docs\"",
+    "build": "turbo run build --filter=\"!docs\" --filter=\"!*-test\" --filter=\"!*-example\"",
+    "build:release": "turbo run build lint test --filter=\"!docs\" --filter=\"!*-test\" --filter=\"!*-example\"",
     "dev": "turbo run dev",
     "format": "prettier --ignore-unknown --cache --check .",
     "format:write": "prettier --ignore-unknown --write .",
     "lint": "turbo run lint",
     "prepare": "husky",
+    "e2e": "turbo run e2e",
     "test": "turbo run test",
     "type-check": "tsc -b --diagnostics",
     "release": "pnpm run check-minor-version && pnpm run build:release && changeset publish",
     "release-snapshot": "pnpm run check-minor-version && pnpm run build:release && changeset publish --tag snapshot",
     "check-minor-version": "node ./scripts/check-minor-version",
-    "update-version": "node ./scripts/update-version",
-    "new-version": "pnpm run build:release && changeset version && pnpm run check-minor-version && pnpm run update-version",
-    "new-snapshot": "pnpm run build:release && changeset version --snapshot && pnpm run update-version"
+    "new-version": "changeset version && pnpm run check-minor-version && pnpm format:write && pnpm run build:release",
+    "new-snapshot": "pnpm run build:release && changeset version --snapshot"
   },
   "devDependencies": {
     "@changesets/cli": "^2.27.1",
-    "eslint": "^8.56.0",
-    "eslint-config-custom": "workspace:*",
-    "husky": "^9.0.10",
+    "@typescript-eslint/eslint-plugin": "^7.8.0",
+    "eslint": "^8.57.0",
+    "eslint-config-next": "^14.2.3",
+    "eslint-config-prettier": "^9.1.0",
+    "eslint-config-turbo": "^1.13.3",
+    "eslint-plugin-react": "7.34.1",
+    "husky": "^9.0.11",
     "lint-staged": "^15.2.2",
+    "madge": "^7.0.0",
     "prettier": "^3.2.5",
     "prettier-plugin-organize-imports": "^3.2.4",
-    "turbo": "^1.12.3",
-    "typescript": "^5.4.3"
+    "turbo": "^1.13.3",
+    "typescript": "^5.4.5"
   },
-  "packageManager": "pnpm@8.15.1",
+  "packageManager": "pnpm@9.0.5",
   "pnpm": {
     "overrides": {
       "trim": "1.0.1",
-      "@babel/traverse": "7.23.2"
+      "@babel/traverse": "7.23.2",
+      "protobufjs": "7.2.6"
     }
   },
   "lint-staged": {

packages/autotool/README.md

@@ -0,0 +1,83 @@
+# @llamaindex/autotool
+
+> Auto transpile your JS function to LLM Agent compatible
+
+## Usage
+
+First, Install the package
+
+```shell
+npm install @llamaindex/autotool
+pnpm add @llamaindex/autotool
+yarn add @llamaindex/autotool
+```
+
+Second, Add the plugin/loader to your configuration:
+
+### Next.js
+
+```javascript
+import { withNext } from "@llamaindex/autotool/next";
+
+/** @type {import('next').NextConfig} */
+const nextConfig = {};
+
+export default withNext(nextConfig);
+```
+
+### Node.js
+
+```shell
+node --import @llamaindex/autotool/node ./path/to/your/script.js
+```
+
+Third, add `"use tool"` on top of your tool file or change to `.tool.ts`.
+
+```typescript
+"use tool";
+
+export function getWeather(city: string) {
+  // ...
+}
+// ...
+```
+
+Finally, export a chat handler function to the frontend using `llamaindex` Agent
+
+```typescript
+"use server";
+
+// imports ...
+
+export async function chatWithAI(message: string): Promise<JSX.Element> {
+  const agent = new OpenAIAgent({
+    tools: convertTools("llamaindex"),
+  });
+  const uiStream = createStreamableUI();
+  agent
+    .chat({
+      stream: true,
+      message,
+    })
+    .then(async (responseStream) => {
+      return responseStream.pipeTo(
+        new WritableStream({
+          start: () => {
+            uiStream.append("\n");
+          },
+          write: async (message) => {
+            uiStream.append(message.response.delta);
+          },
+          close: () => {
+            uiStream.done();
+          },
+        }),
+      );
+    });
+  return uiStream.value;
+}
+```
+
+## License
+
+MIT

packages/autotool/examples/01_node/CHANGELOG.md

@@ -0,0 +1,28 @@
+# @llamaindex/autotool-01-node-example
+
+## null
+
+### Patch Changes
+
+- Updated dependencies [e072c45]
+- Updated dependencies [9e133ac]
+- Updated dependencies [447105a]
+- Updated dependencies [320be3f]
+  - llamaindex@0.3.11
+  - @llamaindex/autotool@0.0.1
+
+## null
+
+### Patch Changes
+
+- Updated dependencies [4aba02e]
+  - llamaindex@0.3.10
+  - @llamaindex/autotool@0.0.1
+
+## null
+
+### Patch Changes
+
+- Updated dependencies [c3747d0]
+  - llamaindex@0.3.9
+  - @llamaindex/autotool@0.0.1

packages/autotool/examples/01_node/package.json

@@ -0,0 +1,17 @@
+{
+  "name": "@llamaindex/autotool-01-node-example",
+  "private": true,
+  "type": "module",
+  "dependencies": {
+    "@llamaindex/autotool": "workspace:*",
+    "llamaindex": "workspace:*",
+    "openai": "^4.43.0"
+  },
+  "devDependencies": {
+    "tsx": "^4.9.3"
+  },
+  "scripts": {
+    "start": "node --import tsx --import @llamaindex/autotool/node ./src/index.ts"
+  },
+  "version": null
+}

packages/autotool/examples/01_node/src/index.tool.ts

@@ -0,0 +1,11 @@
+import { getWeather } from "./utils.js";
+
+/**
+ * Get current location
+ */
+export function getCurrentLocation() {
+  console.log("Getting current location");
+  return "London";
+}
+
+export { getWeather };